Corrective Gene Therapy In A Murine Model Of Familial
Adenomatous Polyposis - A Study of The Efficacy of Gene
Transfer and The Resultant Phenotypic Effects.
A thesis submitted to the University of London for the Degree of Doctor of
Medicine (MD)
By
Rachel Marie Bright-Thomas MA, BM BCh, FRCS
Department of Surgery, The Royal Free and University College Medical
School, University College London
June 2001 ProQuest Number: U643881
All rights reserved
INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest.
ProQuest U643881
Published by ProQuest LLC(2016). Copyright of the Dissertation is held by the Author.
All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code. Microform Edition © ProQuest LLC.
ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 Abstract
The Adenomatous Polyposis Coli {APC) tumour suppressor gene is known to be mutated or deleted in the early stages of the majority of sporadic colorectal cancers, and is thought to initiate neoplasia when its mutation leads to uncontrolled intra-cellular levels of the oncogene p-catenin. Germline mutation of APC gives rise to Familial Adenomatous Polyposis (FAP), a pre-neoplastic syndrome characterised by the development of multiple intestinal polyps, one or more of which invariably becomes malignant. Extra-intestinal manifestations of this disease include potentially fatal fibromatoses (desmoids).
The Min/+ mouse is a murine model of FAP with an homologous germline mutation of Ape. It also develops multiple small intestinal polyps and a smaller number of colorectal tumours. However, it does not develop desmoid tumours.
Corrective gene therapy for cancer has now reached the stage of human clinical trials, most based on replacement of wild type (WT) p53 function in advanced cancers that show mutation or loss of this tumour suppressor. In the case of FAP, it has been postulated that local replacement of WT APC may decrease the incidence or rate of growth of intestinal polyps (and possibly desmoid tumours) by restoration of 'normal’ (3- catenin degradation.
This project studied in vivo transfer of APC into the gut of Min/+ and WT mice. The safety of this treatment was demonstrated, and transgene expression was seen to correlate with a decrease in intestinal P-catenin levels. Intra-peritoneal transfer of APC into WT mice was also performed. This revealed high level transgene expression in tissues known to give rise to desmoid tumours in human disease.
These promising preliminary results in model systems will have to be confirmed in humans, and advances which improve the efficiency of transfection incorporated, before Phase I human trials of^PC gene therapy can be considered. Table of Contents
Title P age...... 1
Abstract...... 2
Table of Contents...... 3
Acknowledgements ...... 18
Statement of Originality...... 19
Abbreviations...... 20
Measurements ...... 25
Chapter 1-Introduction and Historical Review...... 26
1.1 Colorectal Cancer...... 27
1.2 Familial Adenomatous Polyposis...... 29
1.2.1 Incidence and Characterisation of Disease ...... 29
1.2.2 Management of FAP ...... 30
1.2.2.1 Colorectal Disease...... 31
1.2.2.2 Duodenal Disease...... 32
1.2.2.3 Desmoid Disease...... 32
1.2.3 Survival in FAP...... 34
1.3 The APC Gene in FAP...... 34
1.3.1 Genotype-Phenotype Correlation ...... 35
1.3.1.1 Applications to the management of FAP...... 37
1.3.2 'APC Negative’ FAP ...... 37
1.3.3 Somatic APC Mutations in FA P ...... 38
3 1.4 The APC Gene in Sporadic Colorectal Tumours...... 40
1.4.1 APC Polymorphisms ...... 41
1.5 The APC Protein...... 42
1.5.1 Dominant Negative and Dose Dependant Effects of The Protein
Heterodimer ...... 43
1.6 Mechanism of Action of APC...... 44
1.6.1 P-Catenin ...... 44
1.6.1.1 The Role of p-Catenin in Cell-Cell Adhesion and Cell Motility...... 45
1.6.1.2 The Role of p-Catenin In Signal Transduction and Its Interaction With
APC...... 46
1.6.1.3 Are the Two Roles of p Catenin Inter-Related ?...... 47
1.6.1.4 The Importance of p Catenin Alterations in Tumourigenesis...... 48
1.6.1.4.1 The Limited Ability of CTNNBl Mutations To Substitute For APC
Mutations ...... 48
1.6.1.4.2 The Effect Of Qualitative Alterations In p-catenin on Tumour Adhesion
...... 49
1.6.1.4.3 The Effect of Quantitative Alterations In P-Catenin on Tumour Stage
and Grade ...... 50
1.6.2 Other Postulated Functions for APC ...... 51
1.6.2.1 APC and Cell Cycle Progression...... 51
1.6.2.2 APC and Crypt Fission...... 52
1.6.2.3 APC and Cell Migration...... 52
1.6.2.4 APC and DLG/EBl...... 53
1.6.2.5 APC and Apoptosis...... 53
4 1.7 The Interaction of NSAIDS and APC...... 54
1.8 Animal Models Of Colorectal Cancer...... 56
1.8.1 The Min/+ Mouse ...... 56
1.8.1.1 Use of The Min/+ Mouse in Studies of Gene Transfer To Date...... 59
1.8.2 Other Mouse Models of FAP ...... 60
1.8.3 Mouse Models of HNPCC ...... 61
1.9 Gene Therapy...... 62
1.9.1 The History of Human Gene Therapy ...... 63
1.9.2 Gene Delivery ...... 64
1.9.2.1 Viral Vectors...... 64
1.9.2.1.1 Retrovirus ...... 64
1.9.2.1.2 Adenovirus ...... 65
1.9.2.1.3 Adeno-associated virus (AAV)...... 67
1.9.2.1.4 Other Viral Vectors...... 67
1.9.2.2 Non-Viral Vectors...... 67
1.9.2.2.1 Lipofection ...... 6 8
1.9.3 Cancer Gene Therapy...... 69
1.9.3.1 Cytotoxic Gene Therapy...... 70
1.9.3.1.1 Gene Directed Enzyme Prodrug Therapy (GDEPTl ...... 70
1.9.3.1.2 Gene Therapy To Promote The Therapeutic Index of Cytotoxic Drugs 71
1.9.3.2 Immunological Gene Therapy...... 71
1.9.3.2.1 Non-specific Immune Enhancement ...... 72
1.9.3.2.2 Specific Immune Enhancement ...... 73
1.9.3.3 Corrective Gene Therapy...... 73
1.9.3.3.1 Conceptual Problems With Corrective Gene Therapy ...... 73 5 1.9.3.3.2 Antisense ...... 76
1.9.3.3.3 Human Trials of Corrective Gene Therapy ...... 76
1.9.3.3.4 FAP and Sporadic CRC Corrective Gene Therapy ...... 77
1.9.3.3.4.1 Prophylactic Gene Therapy...... 77
1.9.3.3.4.2 Aims o f this Project...... 78
Figure 1-The Adenoma-Carcinoma Sequence...... 79
Figure 2-APC Structure and Function...... 80
Figure 3-Mechanism of Action of APC and p-Catenin...... 81
Figure 4-p Catenin Structure and Function...... 82
Figure 5-The Interaction of APC and NSAIDS...... 83
Chapter 2-Mouse Genotyping...... 84
2.1 Materials and Methods...... 85
2.1.1 DNA Extraction ...... 85
2.1.1.1 Protocol ...... 85
2.1.2 PGR Amplification ...... 86
2.1.2.1 Protocol ...... 87
2.1.3 Gel Electrophoresis of DNA ...... 88
2.1.3.1 Agarose Gel Electrophoresis...... 88
2.1.3.1.1 Preparation of Agarose Gels ...... 89
2.1.3.1.2 Running Agarose Gels ...... 89
2.2 Results...... 90
2.3 Discussion...... 91
Figure 6-Genotyping PCR Reactions...... 92
6 Figure 7-Example of Genotyping A Litter of Four Mice...... 93
Chapter 3-Plasmid Preparation, Purification and Administration 94
3.1 Introduction ...... 95
3.1.1 The Plasmid...... 95
3.1.2 The Animals...... 95
3.2 Materials and Methods...... 96
3.2.1 Transformation of Bacteria ...... 96
3.2.1.1 Protocol for Transformation of Competent Cells...... 96
3.2.2 Assessment of Transformation ...... 97
3.2.3 Amplification of Transformed Cells ...... 98
3.2.3.1 Protocol ...... 98
3.2.4 Purification of Plasmid ...... 99
3.2.4.1 Protocol ...... 99
3.2.5 Spectrophotometric Quantitation of Nucleic Acids ...... 101
3.2.5.1 Protocol ...... 101
3.2.6 Restriction Enzyme Digestion ...... 102
3.2.6.1 Protocol ...... 102
3.2.7 Plasmid /Liposome Complex Formation ...... 103
3.2.7.1 Protocol ...... 103
3.2.8 Rectal Treatment ...... 103
3.2.8.1 Protocol...... 103
3.2.9 Gavage Treatment ...... 104
3.2.9.1 Protocol...... 105 7 3.2.10 Intra-peritoneal Treatment ...... 106
3.2.10.1 Protocol ...... 106
3.3 Results...... 107
3.3.1 Plasmid Preparation ...... 107
3.3.2 Restriction Enzyme Digestion ...... 107
3.3.3 Plasmid Administration...... 107
3.4 Discussion...... 108
3.4.1 Plasmid Preparation ...... 108
3.4.2 Restriction Enzyme Digestion ...... 108
Figure 8-The APC pCMVP-Neo-Bam Plasmid...... 109
Figure 9-Restriction Enzyme Digestion of Purified APC pCMVp-Neo-
B am...... 110
Chapter 4-Study of Transgene Expression Following APC Gene
Therapy...... I l l
4.1 Methods...... 112
4.1.1 Organ Retrieval ...... 112
4.1.2 Preparation of mRNA ...... 113
4.1.2.1 Protocol ...... 114
114
4.1.2.1.2 Nucleic Acid Extraction ...... 115
4.1.2.1.3 Binding ...... 115
4 . 1 .2 .1.4 Washing ...... 116 4.1.2.1.5 Elution Step ...... 116
4.1.2.1.6 Concentration of mRNA ...... 116
4.1.3 Preparation of cDNA ...... 117
4.1.3.1 Protocol ...... 117
4.1.4 Amplification of cDNA by The Polymerase Chain Reaction 118
4.1.4.1 Protocol ...... 120
4.2 Results...... 121
4.2.1 Transgene Expression Following Rectal Treatment ...... 121
4.2.1.1 Recto-sigmoid Samples...... 121
4.2.1.2 Colonic samples...... 122
4.2.1.3 Liver and Gonads...... 122
4.2.1.4 Controls...... 122
4.2.2 Transgene Expression Following Gavage ...... 123
4.2.2.1 Stomach and Small Bowel Samples...... 123
4.2.2 2 Liver and Gonads...... 123
4.2.2 3 Controls...... 123
4.2.3 Transgene Expression Following Intra-peritoneal Treatment 123
4.2.3.1 Peritoneal and Mesenteric Samples...... 123
4.2.3.2 Intestinal Tissues...... 124
4.2.3.3 Liver and Gonads...... 124
4.2.3 4 Controls...... 124
4.3 Discussion...... 125
4.3.1 Preparation of mRNA ...... 125
4.3.2 Preparation of cDNA ...... 125
4.3.3 Amplification of cDNA by the Polymerase Chain Reaction ...... 125 9 4.3.3.1 Primers For Amplification of APC pCMVp-Neo-Bam...... 125
4.3.4 Transgene Expression Following Rectal Treatments ...... 126
4.3.4.1 Recto-sigmoid Samples...... 126
4.3 4.2 Colonic Samples...... 127
4.3.4 3 Liver and Gonads...... 127
4.3.5 Transgene Expression Following Gavage ...... 128
4.3.5.1 Stomach and Small Bowel Samples...... 128
4.3.5.2 Liver and Gonads...... 129
4.3.6 Transgene Expression Following Intra-peritoneal Treatment 130
4.3.6.1 Peritoneal and Mesenteric Samples...... 131
4.3.6.2 Intestinal Tissues...... 132
4.3.6.3 Liver and Gonads...... 132
Tables 1 and 2-Protocols For Animal Euthanasia...... 134
Figure 10- Agarose Gel Electrophoresis of rtPCR Products From Recto-
Sigmoid Tissue of All Rectally Treated Mice...... 135
Figure 11- Agarose Gel Electrophoresis of rtPCR Products From
Colonic Tissue of Rectally Treated Mice Which Expressed The
Transgene In The Rectum...... 136
Figure 12- Agarose Gel Electrophoresis of rtPCR Products From The
Livers of Mice Treated With Rectal APC pCMVP-Neo-Bam...... 137
Figure 13- Transgene Expression Following Rectal Treatment With
APC pCMVP-Neo-Bam...... 138
Figure 14- Agarose Gel Electrophoresis of rtPCR Products From The
Stomach of All Gavage Treated Mice...... 139 10 Figure 15- Agarose Gel Electrophoresis of rtPCR Products From The
Proximal Small Bowel of All Gavage Treated M ice...... 140
Figure 16- Agarose Gel Electrophoresis of rtPCR Products From The
Livers of All Gavage Treated Mice...... 141
Figure 17- Agarose Gel Electrophoresis of rtPCR Products From The
Gonads of All Gavage Treated M ice...... 142
Figure 18- Transgene Expression Following Gavage With APC pCMVP-
Neo-Bam...... 143
Figure 19- Agarose Gel Electrophoresis of rtPCR Products From
Peritoneal and Mesenteric Samples of All Mice treated via The
Intra-peritoneal Route...... 144
Figure 20- Transgene Expression Following Intraperitoneal Treatment
With APC pCMVP-Neo-Bam...... 145
Table 3- Sites of Transgene Expression In Individual Animals After
Intraperitoneal Treatment With APC pCMVP-Neo-Bam...... 146
Chapter 5-Clinical Effects of^PC Gene Therapy...... 147
5.1 Materials and Methods...... 148
5.2 Results...... 148
5.2.1 Rectal Trial...... 148
5.2.2 Gavage Trial ...... 149
5.2.3 Intra-peritoneal T rial ...... 150
11 5.3 Discussion...... 150
5.3.1 Unwanted Systemic Effects of Treatment ...... 150
5.3.2 Beneficial Effect of Treatment on Polyp Number ...... 151
5.3.2.1 Rectal Trial...... 152
5.3.2.2 Gavage Trial...... 152
Figure 21-Polyp Count In The Intestine Of Individual Min/+ Mice
Treated With Rectal APC pCMVp-Neo-Bam or ’LIPOFECTAMINE^^
Alone* ...... 155
Figure 22- Comparison of Median Polyp Load in The Large Intestine of
Min/+ Mice Treated With Rectal APC pCMVP-Neo-Bam or
'LIPOFECTAMINE™ Alone'...... 156
Figure 23-Polyp Count In The Intestine of Individual Min/+ Mice
Treated By Gavage With APC pCMVp-Neo-Bam or
'LIPOFECTAMINE™ Alone'...... 157
Figure 24- Comparison of Median Polyp Load in The Small Intestine of
Min/4- Mice Treated By Gavage With APC pCMVP-Neo-Bam or
'LIPOFECTAMINE™ Alone'...... 158
Chapter 6-Protein Alterations Occurring After ^PC Gene Therapy.. 159
6.1 Introduction ...... 160
6.2 Materials and Methods...... 160
6.2.1 Western Blotting ...... 160
6.2.1.1 Preparation of the Protein Sample...... 161 12 6.2.1.1.1 Apc...... 162
6.2.1.1.2 p-Catenin...... 162
6.2.1.2 Quantitation of Protein Concentration...... 162
6.2.1.2.1 Protocol ...... 163
6.2.1.3 Denaturing Gel Electrophoresis...... 164
6.2.1.3.1 Ape...... 165
6.2.1.3.2 P-Catenin ...... 165
6.2.1.3.2.1 Protocol ...... 165
6.2.1.4 Staining of Protein Gels...... 166
6.2.1.4.1 Coomassie Brilliant Blue Staining Protocol ...... 167
6.2.1.4.2 Silver Staining ...... 167
6.2.1.4.2.1 Protocol ...... 168
6.2.1.5 Transfer To A Membrane Support...... 169
6.2.1.5.1 Ape...... 169
6.2.1.5.1.1 Protocol ...... 169
6.2.1.5.2 P-catenin ...... 171
6.2.1.5.2.1 Protocol ...... 171
6.2.1.6 Staining of Proteins Covalently Bound To The Nitrocellulose Membrane
...... 172
6.2.1.6.1 Ponceau S Staining Protocol ...... 172
6.2.1.6.2 India Ink Staining Protocol ...... 173
6.2.1.7 Blocking Non-Specific Binding Sites On The Membrane ...... 173
6.2.1.7.1 Protocol ...... 174
6.2.1.8 Addition Of Primary and Secondary Antibodies...... 174
6.2.1.8.1 Protocol ...... 174
6.2.1.9 Detection of Signal...... 175
13 6.2.1.9.1 Protocol ...... 176
6.2.1.9.2 Signal Calibration ...... 177
6.2.1.10 Stripping and Re-probing Membrane...... 177
6.2.1.10.1 Protocol ...... 177
6.2.1.11 Densitometry and Statistical Analysis Of Results...... 178
6.2.2 Immunohistochemistry...... 179
6.2.2.1 Protocol ...... 180
6.3 Results ...... 181
6.3.1 Untreated Animals ...... 181
6.3.1.1 Western Blotting ...... 181
6.3.1.1.1 APC...... 182
6.3.1.1.2 p-Catenin ...... 183
6.3.1.2 Immunohistochemistry...... 184
6.3.1.2.1 Ape...... 184
6.3.1.2.2 p-catenin ...... 184
6.3.2 Treated Animals ...... 185
6.3.2.1 Rectal Trial...... 185
6.3.2.2 Gavage Trial...... 186
6.4 Discussion ...... 187
6.4.1 Untreated Animals ...... 187
6.4.1.1 Western Blotting ...... 187
6.4.1.1.1 APC...... 187
6.4.1.1.2 B-catenin ...... 189
6.4.1.1.2.1 Intestinal Results...... 190
6.4.1.2 Immunohistochemistry...... 192
14 6.4.2 Treated Animals ...... 192
6.4.2.1 Rectal Trial...... 193
6.4.2 2 Gavage Trial...... 194
Figure 25-Estimation of Protein Concentration...... 196
Figure 26- Immunblots of Recto-Sigmoid Tissue Samples Taken From
Untreated Mice...... 197
Figure 27- Statistical Comparison of P-catenin Levels in The Recto-
Sigmoid of Untreated Min/+ and WT M ice...... 198
Figure 28- Immunblots of Proximal Small Bowel Tissue Samples Taken
From Untreated Mice...... 199
Figure 29- Statistical Comparison of Total and Sub-cellular Levels of P- catenin in The Proximal Small Intestine of Min/+ and WT Mice 200
Figure 30- Haematoxylin and Eosin Staining of Min/+ Intestinal Frozen
Sections...... 201
Figure 31-Immunoblotting APC in Intestinal Frozen Sections 202
Figure 32-Immunoblotting p Catenin in Intestinal Frozen Sections... 203
Figure 33-Immunblots of Recto-Sigmoid Tissue Samples Taken From
Rectally treated Mice...... 204
Figure 34- Alterations in Recto-Sigmoid Levels of p-catenin In Response
To Transgene Expression...... 205
Figure 35- Alterations in Recto-Sigmoid Levels of P-catenin In Response
To Treatment with APC pCMVP-Neo-Bam orLIPOFECTAMINE^^ . 206
15 Figure 36-Immunblots of Tissue Samples Taken From The Proximal
Small Bowel of Mice Treated By Gavage...... 207
Figure 37- Alterations in Proximal Small Bowel Levels of p-catenin In
Response To Transgene Expression...... 208
Figure 38- Alterations in Levels of P-catenin In The Proximal Small
Bowel In Response To Treatment with APC pCMVp-Neo-Bam or
LIPOFECTAMINE™...... 209
Chapter 7-Discussion...... 210
7.1 Rationale Underlying Study Design...... 211
7.1.1 Liposome Mediated Gene Therapy ...... 211
7.2 Summary of Experimental Findings...... 212
7.2.1 Rectal Administration of APC pCMVp-Neo-Bam ...... 212
7.2.2 Administration of APC pCMVp-Neo-Bam By Gavage ...... 213
7.2.3 Intra-peritoneal Administration of APC pCMVp-Neo-Bam ...... 215
7.3 Limitations of The Study...... 216
7.4 The Future of Corrective Gene Therapy...... 216
7.4.1 Viral Vectors ...... 217
7.4.2 Non-Viral Vectors ...... 218
7.4.3 Hybrid Vectors...... 219
7.4.4 Targeting Vectors ...... 220
7.4.5 New Vectors ...... 222
16 7.4.6 New Forms of Gene Therapy ...... 223
7.4.6.1 Antiangiogenic gene therapy...... 223
7.4.6 2 Targeted Gene Repair...... 224
7.5 The Future Role of APC Gene Therapy In Colorectal Cancer 225
Chapter 8-Appendices...... 227
Appendix 1-Mouse Genotyping...... 228
Appendix 2- Agarose Gel Electrophoresis...... 229
Appendix 3-Plasmid Preparation and Purification...... 230
Appendix 4-Preparation of Messenger RNA...... 233
Appendix 5- Reverse Transcriptase PCR...... 235
Appendix 6-Preparation of The Protein Sample...... 237
Appendix 7-Denaturing Gel Electrophoresis...... 240
Appendix 8- Staining of Protein Gels...... 242
Appendix 9-Transfer To A Membrane Support...... 243
Appendix 10- Staining of Membranes...... 245
Appendix 11-lmm unoblotting ...... 246
Appendix 12- Immunohistochemistry...... 248
Chapter 9-References...... 250
17 Acknowledgements
This project was initially supported by the Royal College of Surgeons of England through the 'Charles Rob Research Fellowship' and then by a joint research training fellowship from the Alimentary Pharmacology and Therapeutics Trust and Digestive Disorders
Foundation. I am extremely grateful to all three organisations.
I wish to thank my supervisors, Professor Irving Taylor and Miss Rachel Hargest for their
inspiration and support. Dr Harpreet Wasan of the Imperial Cancer Research Fund allowed me access to the mouse model. He, Dr Alan Mackay, Dr Rob Harris, Dr Sioban
Sengupta, Dr Rami Suzuki and Dr Felicity Savage, of University College London, provided invaluable advice regarding technical aspects of the work.
I also wish to thank Del Watling, Rob Rudlin, Julie Bee, and all the animal technicians who made the animal work run smoothly.
Finally I would like to thank my husband, mother and all of my family for their endless
support and encouragement over the last two years.
18 Statement of Originality
The studies described and presented in this thesis are the original work of the author. All molecular biology techniques were performed personally in the Department of Surgery,
University College London.
Animal studies were performed under the appropriate Home Office project and personal licences and with ethical committee approval from both University College London and the Imperial Cancer Research Fund (ICRF). All animal treatments were performed by the author, with the exception of gavage of small mice. This was performed by ICRF animal house technicians, in the presence of the author. The animal technicians also provided day to day care of the mice used in this project.
No part of this work has been submitted to any other university for consideration for a higher degree.
19 Abbreviations
A Adenosine
AAPC Attenuated Adenomatous Polyposis
AAV Adeno-associated virus
Ab Antibody
ACF Aberrant crypt foci
ADA Adenosine deaminase
Ag Antigen
AgNOs Silver nitrate
APC Adenomatous Polyposis Goli protein
APC Human Adenomatous Polyposis Goli gene
Ape Murine Adenomatous Polyposis Goli gene
APS Ammonium persulphate
BAG Bacterial artificial chromosome
EGA Eicinchoninic acid bp Ease pair
ESA Bovine serum albumin
P-gal p-D-galactosidase
"G Degrees Celsius
G Cysteine
GAR Goxsackie adenoviral receptor
GEA Carcinogenic embryonic antigen
GHRPE Congenital hypertrophy of the retinal pigment epithelium
GMV Cytomegalovirus
GOX-2 Gyclooxygenase 2
GpG Cytosine phospho-guanine
20 CRC Colorectal cancer
CT Computed tomography
CTNNBl P-catenin gene
CTNNGl y-catenin gene
DAB Diaminobenzidine
DCC Deleted In Colorectal Cancer gene ddHiO Double distilled water
DEPC Diethyl pyrocarbonate
DLG Drosophila discs large tumour suppressor protein
DNA Deoxyribonucleic acid dNTP Deoxynucleotide tri-phosphate
DOPE Dioleoylphosphatidylethanolamine ds Double stranded
DTT Dithiothreitol
E. Coli Escherichia Coli
EDTA Ethylene diamine tetraacetic acid
EGF-R Epidermal growth factor receptor
EtBr Ethidium bromide
FAP Familial Adenomatous Polyposis
Fc Antibody constant fraction g Gravity
G Guanine
GDEPT Gene directed enzyme prodrug therapy
GI Gastro-intestinal
GTAC Gene Therapy Advisory Committee
GTC Guanidinium thiocyanate
21 HAC Human artificial chromosome
H&E Haematoxylin and eosin
HCl Hydrochloric acid
Her-2/neu Q-erb B-2 growth factor receptor
HGF-R Hepatocyte growth factor receptor mv-1 Human immunodeficiency virus type 1
HNPCC Hereditary Nonpolyposis Colorectal Cancer
H 2 O Water
H 2 O2 Hydrogen peroxide
HRPO Horseradish peroxidase
HSV Herpes simples virus
ICRF Imperial Cancer Research Fund
Ig Immunoglobulin
IGFIIR Insulin-like growth factor II receptor gene
IRA I leo-rectal anastomosis
Kb Kilobase kD Kilodaltons
K3 +EDTA Potassium ethylene diamine tetraacetic acid
LB Luria-Bertani broth
LOH Loss of heterozygosity
MAC Mammalian artificial chromosome
MAMA Monoallelic mutation analysis
MCC Mutated In Colorectal Cancer gene
MCR Mutation Cluster Region
MDCK Madin-Darby canine kidney
MDR-1 Multidrug resistance gene
22 Mg CI 2 Magnesium chloride
MHC Major histocompatability complex
MIN Multiple Intestinal Neoplasia
MMR Mismatch repair
Mom-1 Modifier of Min 1 gene
MOPS 3[N-Morpholino]propanesulfonic acid
MSH2 Mut S homologue 2 (human mismatch repair gene)
MSI Micro satellite instability
Na Cl Sodium chloride
Na OH Sodium hydroxide
Neo Neomycin
NP-40 Nonidet-P40
NSAID Non-steroidal anti-inflammatory drug
NSCLC Non small cell lung cancer
ODN Oligodeoxynucleotide
PAGE Polyacrylamide gel electrophoresis
PBS Phosphate buffered saline
PCR Polymerase chain reaction
PEG Polyethylene glycol
PF4 Platelet factor 4
PI3 -K Pho spho ino sitide3 -kinase
PINC Protective interactive and non-condensing polymer
PIPES Piperazine-NN'-bis-2-ethanesulphonic acid
Pla2g2a Secreted phospholipase 2A gene
PMSF Phenylmethylsulfonylfluoride
PPARÔ Peroxisome proliferator-activated receptor ô
23 PPC Pan-proctocolectomy
Ptgs2 COX-2 gene
PTP Protein-tyrosine phosphatase
RAC Recombinant DNA Advisory Committee
RCA Replication competent adenovirus
RER Replication error
RNA Ribonucleic acid
RPC Restorative proctocolectomy
SCCHN Squamous cell carcinoma of the head and neck
SCID Severe combined immunodeficiency
SDS Sodium dodecyl sulphate
SPLP Stabilised plasmid-lipid particles ss Single stranded
SSCP Single strand conformational polymorphism
T Thymidine
TAE Tris-acetate
TB Terrific broth
TBE Tris-borate
TBS Tris buffered saline
TE Tris-EDTA
TEMED N,N,N',N'-tetramethylethylenediamine
TGFBR2 Transforming growth factor-^ type II receptor gene tk Thymidine kinase
Tm Melting temperature
TPE Tris-phosphate
Tris Tris (hydroxymethyl) amino methane
24 Tris-HCl Tris (hydroxymethyl) methyl ammonium chloride
UKCCCR United Kingdom Co-ordinating Committee on Cancer Research
UV Ultra-violet
VEGF Vascular endothelial growth factor
V Volts
W Watts
WT Wild-type
X-gal 5-bromo-4-chloro-3-indolyl-P-D-galactoside
YAC Yeast artificial chromosome
Measurements
gram M Molar mg milligram mM millimolar
Pg microgram pM micromolar ng nanogram nM nano molar
Pg picogram pM picomolar
litre cm centimetre ml millilitre mm millimetre
Hi microlitre
25 Chapter 1
Introduction and Historical Review
26 This project investigates the in vivo efficiency and clinical and molecular effects of transfer of the human tumour suppressor gene Adenomatous Polyposis Coli (APC) into an animal model of Familial Adenomatous Polyposis (FAP). FAP is a pre-neoplastic condition caused by germline mutation or loss of the APC gene. In the absence of prophylactic surgery it invariably gives rise to colorectal cancer. FAP is also widely used as a model for the genetic changes underlying sporadic colorectal carcinogenesis.
The introductory chapter gives an overview of the genetic basis of colorectal carcinoma, focusing largely on FAP as a disease model. The APC gene and its protein product are then discussed along with possible mechanisms of action, p-catenin, the protein that appears to be an important effector of APC mutations, is the main focus of this section.
The Min/+ mouse model of FAP is then described along with other murine models of colorectal cancer. Finally the basic concepts of gene therapy are covered, concentrating
on lipofection, the route of administration used in this project.
1.1 Colorectal Cancer
Colorectal cancer (CRC) is an extremely common cancer. The most recent international
figures show that over 783,000 new cases were diagnosed in 1990, and the annual death
rate attributable to CRC is approximately 394,000 (Boyle and Langman, 2000). If
diagnosed in the early stages tumours may be removed surgically with minimal morbidity
and mortality. However, presentation with late stage disease is common and the prognosis
for these patients has hardly altered over the last 30 years with only a 37% overall 5 year
survival (Black et al., 1993). This makes CRC the second largest cause of death due to
cancer after lung cancer. Improvements in chemotherapy, radiotherapy or surgery to date
have done little to affect these statistics, but it is hoped that a better understanding of the
underlying disease processes will lead to more effective intervention.
27 There are at least two paths by which a 'normal' colonic epithelial cell can enter the neoplastic process. The most common route or 'classical' route is the adenoma-carcinoma sequence where loss or mutation of the tumour suppressor gene Adenomatous Polyposis
Coli {APC) initiates the neoplastic process, and it gathers pace with the accumulation of other mutations in vital proto-oncogenes including K-ras, deleted in colorectal cancer
(DCC) and p53 (Fearon and Vogelstein, 1990) (See Figure 1). These tumours are invariably aneuploid with gross chromosomal abnormalities and loss of many randomly chosen alleles (Aaltonen et al., 1993, Bocker et al., 1996). The rate-limiting step along the pathway is probably the initiation of the adenoma.
In contrast, 10-15% of CRCs have a grossly normal cytogenetic profile but on closer inspection they demonstrate microsatellite instability (MSI), a marker for generalised genetic instability. They acquire mutations in the same oncogenes but in a more subtle way, generally through errors of replication (Aaltonen et al., 1993, Bocker et al., 1996,
Huang et al., 1996). Once the process has started these replication error positive (RER+) tumours appear to acquire new mutations at an increased rate and move rapidly jfrom an adenoma to a carcinoma.
There are two major inherited forms of CRC, Familial Adenomatous Polyposis (FAP) and
Hereditary Nonpolyposis Colorectal Cancer (HNPCC), accounting for 1% and 6 % o f all
colorectal cancers respectively. They are both autosomal dominantly inherited disorders
but the underlying genetic abnormalities are very different. FAP is due to a germline mutation in the APC gene and HNPCC is linked to germline mutations in mismatch repair
(MMR) genes, important in maintaining fidelity in the replication of DNA. Although it is
an oversimplification, FAP could be considered as disease of tumour initiation whereas
HNPCC leads to an accelerated rate of tumour progression once an initiating event has
28 occurred. In depth study of these genetically defined disorders should eventually provide a better understanding of the pathogenesis of the majority of sporadic CRC and prophylactic therapy initially aimed at FAP or HNPCC kindreds may be applicable to others at increased risk of colorectal cancer.
1.2 Familial Adenomatous Polyposis
1.2.1 Incidence and Characterisation of Disease
According to the Danish FAP Register, the mean annual incidence of FAP was 1.85 per million taken over the period 1971 to 1992, with a prevalence of 32 per million by the end of this period (Bulow et al., 1996). The completeness of registration in Denmark is remarkably high, 97%, so these are likely to be true population statistics for Western
Europe. FAP affects both sexes equally and is characterised by the development of hundreds to thousands of adenomatous polyps throughout the colon and rectum in the second and third decades of life. The diagnosis is confirmed by histological identification of at least 100 adenomatous polyps (Bussey, 1975), although attenuated forms of disease with a reduced level of polyposis in the presence of a causative germline mutation are now recognised (Spirio et al., 1993).
Once initiated, each adenoma progresses individually along the adenoma-carcinoma sequence (Fearon and Vogelstein, 1990) (See Figure 1). Without treatment an affected individual will inevitably develop at least one colorectal carcinoma. A small proportion of affected families (approximately 10%) have an attenuated or atypical form of the disease where the colonic polyp load is decreased and the progression to malignant disease in the large bowel occurs at a later age (Spirio et al., 1993, Philips et al., 1994, Campbell et al.,
1994).
29 However, FAP is not just a pre-neoplastic disease of the colon but a genetically defined multitumoural syndrome. Upper gastrointestinal tract polyps are common. Fundic gland polyps (seen in 40% of patients) are generally benign hamartomas, but antral and duodenal adenomas (affecting 5-10% or 90% of patients respectively) are adenomatous polyps that carry a risk of progression to frank malignancy (Spigelman et al., 1989).
Indeed, following prophylactic removal of the colon, peri-ampullary carcinoma is one of the two commonest causes of death in FAP (Arvanitis et al., 1990).
Extraintestinal manifestations of disease include osteomas (75%), congenital hypertrophy
of the retinal pigment epithelium (CHRPE) (75%), epidermoid cysts (50%), dental
anomalies (17%) and desmoid tumours (10-15%) (Regillo et al., 1993, Clark and Phillips,
1996, Gurbuz et al., 1994). Desmoid tumours (described below) are the other leading
cause of morbidity and mortality in patients with FAP following colectomy (Arvanitis et
al., 1990). Rare conditions are also described in association with FAP, including brain
tumours (Turcot et al., 1959), hepatoblastoma (Giardiello et al., 1992) and papillary
carcinoma of the thyroid (Plail et al., 1987). Although uncommon these too can be fatal in
affected individuals.
1.2.2 Management of FAP
In order to address controversial issues in the management of FAP, the Leeds Castle
Polyposis Group was set up in 1985 (Northover, 1987). This is an international
organisation made up of specialists with a particular interest in FAP, including surgeons,
gastroenterologists and geneticists. The group meets biennially to discuss research and
formulate management plans.
30 At a national or regional level the care of subjects with FAP and their families is generally organised within the context of a register of affected kindreds. Pedigrees are drawn up around an index case and treatment protocols are set out for affected patients as appropriate.
1.2.2.1 Colorectal Disease
The treatment of choice of the colon and rectum in FAP is prophylactic surgery between the ages of 15 and 25 years. There are three main surgical options; a sub-total colectomy with an ileorectal anastomosis (IRA), a pan-proctocolectomy with a terminal ileostomy
(PPG), or a restorative proctocolectomy with formation of an ileo-anal pouch (RPC).
Opinion has been divided as to best course of action but the majority of surgeons favour an IRA in the first instance (Vasen et al., 1999).
An IRA is a single stage procedure with a low complication rate and negligible effect on continence in a previously healthy group of young patients. However, it does leave ’at risk' rectal mucosa which needs to be kept under surveillance and may require further surgery at a later date. In contrast, RPC is often a two or three stage procedure with a higher rate of complications, including permanent impairment of continence, intestinal obstruction and pouch sepsis or dysfunction. Any of these can ultimately necessitate removal of the pouch and formation of a permanent ileostomy (Keighley et al., 1993).
PPC is still performed when it is necessary to remove the rectum but it is undesirable to restore continuity.
Where IRA is the primary operation, patients have an overall cumulative risk of approximately 30% of developing rectal carcinoma by the age of 60 years (Nugent and
Phillips, 1992). Therefore regular endoscopic surveillance of the rectum is performed at
31 3-12 monthly intervals, dependant upon the number, size and level of dysplasia of the polyps. Disease progression signals the need for further surgery in the form of a semi- elective proctectomy and either ileo-anal pouch formation or creation of a terminal ileostomy. At the moment non-steroidal anti-inflammatory drugs (NSAIDS) are the only usefiil adjuvant treatment for patients with enlarging rectal polyps (Nugent et al., 1993a,
Steinbach et al., 2000). The possibility of adding in corrective gene therapy to further reduce the rate or incidence of polyp progression is tantalising.
1.2.2.2 Duodenal Disease
The management of upper GI polyps is a controversial issue. Local ablation with laser or argon plasma coagulation therapy is used. However there is no firm evidence that these confer a survival benefit. Surgery is a significant undertaking, traditionally associated
with considerable morbidity. The options are a pancreas-sparing duodenectomy or a pylorus preserving pancreatico-duodenectomy (Vasen et al., 1999).
In less aggressive disease NSAIDS have been shown in clinical trials to reduce duodenal
epithelial cell proliferation and encourage polyp regression (Nugent et al., 1993a).
However, any additional prophylactic or therapeutic intervention to deter polyp
progression would be beneficial. This is another potential role for APC corrective gene
therapy in FAP.
1.2.2.3 Desmoid Disease
Desmoid tumours are benign fibromatoses arising in musculoaponeurotic tissues. They
can be aggressive in their infiltration and compression of surrounding structures, leading
to considerable morbidity and even mortality. Desmoid disease is unusual in the general
population but is a well-recognised extra-intestinal manifestation of FAP, affecting
32 approximately 10% of patients, with subclinical disease present in an even greater percentage (Clark and Phillips, 1996). The majority of desmoids in FAP are intra- abdominal (70%), and of these 40-50% arise in the mesentery (Einstein et al., 1991,
Soravia et al., 2000)
Desmoids raise possibly the most difficult management decisions in FAP. Surgery provides temporary alleviation of local symptoms, especially for abdominal wall tumours.
However, intra-abdominal surgery for desmoids is hazardous and is thought to be an initiating factor in their development (Clark et al., 1999, Soravia et al., 2000). Thus it is generally reserved for relief of obstruction or treatment of perforation. Radiotherapy may be beneficial in decreasing post-operative recurrence rates, especially in extra-abdominal disease (Nuyttens et al., 2000, Ballo et al., 1999). However, radio sensitivity of the intestinal tract has limited its use for intra-abdominal desmoids. Medical treatment with
NSAIDS or oestrogenic compounds (such as tamoxifen) may be beneficial (Wilcken and
Tattersall, 1991, Tsukada et al., 1992) and in some cases advanced disease has been effectively treated with combination cytotoxic chemotherapy (Skapek et al., 1998, Patel et al., 1993). However, chemotherapy has significant side effects without a guarantee of cure or remission.
Since it has been demonstrated that desmoids in FAP, like the intestinal adenomas, are clonal neoplasms arising from germline mutation or loss of one APC allele and somatic mutation or loss of the second allele (Dangel et al., 1994, Palmirotta et al, 1995, Miyaki et al, 1993), it is possible that corrective gene therapy could offer a new form of prophylaxis and/or treatment.
33 1.2.3 Survival in FAP
In a single centre study, survival of 161 FAP patients diagnosed with colorectal
carcinoma was compared to that of 2,035 patients with sporadic colorectal cancer matched for age, sex, stage and tumour location. There was no statistically significant
difference in survival between the two groups (Bertario et al., 1999). Other studies have
confirmed this result (Aarnio et al., 1998). This seems logical as the inherited APC mutation predisposes to adenoma formation at an early age but subsequent events appear to proceed in a similar fashion to those recorded in sporadic disease.
However, CRC is not the only cause of early death in individuals with FAP. Indeed,
retrospective life table analysis of over 200 individuals with FAP (who had undergone
IRA) demonstrated a relative mortality rate in excess of 3 when compared to the general
population (Nugent et al., 1993b). The three main causes of death were upper GI
malignancy, desmoids and perioperative complications.
1.3 The APC Gene in FAP
The^PC gene was identified and sequenced in 1991 (Nishisho et al., 1991, Kinzler et al.,
1991, Groden et al., 1991). It is situated at Chromosome 5q21-22. The gene is large,
being composed of approximately 100,000 bp. Following transcription, the mRNA is still
almost 10,000 bp long and made up of 15 exons (Groden et al., 1991). The largest of
these is exon 15, corresponding to codons 900 to 1600. Over 300 different germline
mutations have been identified in FAP kindreds to date (Beroud and Soussi, 1996) (see
Figure 2). The majority of clinically significant mutations have been localised to codons
1055-1309 in the 5' half of exon 15 (Miyoshi et al., 1992a, Nagase and Nakamura, 1993),
although classical FAP is known to be associated with mutations as far 5' as codons 168
to 170 in exon 4 (Olschwang et al., 1993a, Fodde et al., 1992). Germline mutations
34 occurring in more 5' and 3’ positions within the gene are less common and often associated with milder colonic disease (as discussed below).
Studies have demonstrated that the germline mutations underlying FAP are generally short insertions, deletions or single nucleotide substitutions in the APC gene. The majority of the single base changes occur at cysteine (C) residues, when C is converted to another base. These mutations tend to produce an inappropriate stop codon leading to premature truncation of the protein product (Nagase and Nakamura, 1993). To date, the most common mutations responsible for classical FAP are a 5bp deletion at codon 1309
(5-9%), a 5bp deletion at codon 1061 (5%), and a 4bp deletion at codon 1068 (2.5%)
(Mandl et al., 1994, Nagase and Nakamura, 1993). However, novel mutations continue to be identified (Monakov et al., 1998, Delatycki et al., 1998, O'Sullivan et al., 1998) and it is becoming increasingly clear that even the fi*equency of common mutations varies with the population studied (Schnitzler et al, 1998).
1.3.1 Genotype-Phenotype Correlation
Phenotypic heterogeneity is seen both between and within FAP kindreds. The strongest genotype-phenotype correlation is seen with classical FAP mutations, for example mutations at codon 1309 are almost always characterised by aggressive disease, with the development of a large number of polyps at a young age (Caspari et al, 1994, Mandl et
a l, 1994).
In contrast, other mutations give rise to more variable extracolonic manifestations of
disease with a relative oligopolyposis. Mutations 5' of codon 157 in exon 4 were the first to be described as giving rise to ‘attenuated adenomatous polyposis’ (AAPC) with under
100 colonic polyps and a late onset of colorectal disease (Spirio et al, 1993). It rapidly
35 became clear that, despite identical germline mutations, the number of colonic polyps and the incidence and severity of other tumours showed a striking variation within each
AAPC kindred (Giardiello et al., 1997). Similarly, although relatively few mutations have been recorded in the most 3' third of the APC gene, codons 1896 to 2844 (Beroud and
Soussi, 1996), those which are recognised also give rise to an attenuated polyposis phenotype with variable expression of extra-intestinal features (Friedl et al., 1996, van der
Luijt et al., 1996, Brensinger et al., 1998, Gardner et al., 1997). Occasional constitutional mutations in the alternatively spliced portion of exon 9 are another group of germline mutations known to give rise to less florid colorectal disease (van der Luijt et al., 1995).
The reason for colonic disease attenuation in all of these mutations appears to lie in their effect on the protein product of the APC gene (as discussed below).
The retinal lesion CHRPE, is associated with truncating mutations stretching from codon
413 in exon 9 upto codon 1387 or 1444 in exon 15. More 5' mutations are not associated with CHRPE (Olschwang et al., 1993b, Davies et al., 1995) (see Figure 2).
Desmoid disease is strongly associated with mutations between codons 1445 and 1578, i.e. in the region just 3' to that associated with retinal abnormalities. Whereas the overall prevalence of desmoids is 4-13% in patients with FAP, almost all patients with mutations within this region have early onset of severe desmoid disease and associated osteomas, epidermoid cysts and upper gastrointestinal polyps (Caspari et al., 1995, Davies et al.,
1995), a combination of features sometimes known as Gardner Syndrome (Gardner and
RC, 1953, Gardner and Plenk, 1952).
36 1.3.1.1 Applications to the management of FAP
Reliable genotype-phenotype correlation, allowing accurate disease prediction, would be usefiil to guide specific patient surveillance and treatment protocols. At the moment this is not possible, but certain guidelines have been suggested. For example, colonoscopic rather than sigmoidoscopic screening is beneficial in kindreds with non-classical 5’ or 3’ mutations where there is likely to be an increased frequency of oligopolyposis and rectal
sparing (Burt et al., 1995). Subjects within these families may also benefit fi*om upper gastrointestinal surveillance at an early stage, even before the development of colorectal polyposis.
1.3.2 ’APC Negative' FAP
Standard screening methods (single strand conformational polymorphism, direct DNA
sequencing and the protein truncation test) have failed to detect an APC mutation in
approximately 20% of kindreds clinically affected by FAP (Heinimann et al., 1998, van
der Luijt et al, 1997). Probands within these 'APC negative' families are generally
diagnosed at a greater age than their 'APC positive' counterparts and have a less severe
phenotype, similar to that of AAPC (Spirio et al, 1993, Heinimann et al, 1998).
However, monoallelic mutation analysis (MAMA) in these patients, revealed that the
majority have significantly reduced expression fi’om one APC allele (Laken et a l, 1999).
In the small group of individuals with FAP who still appear to be truly 'APC negative' it is
likely that gene mutations at unidentified distant loci produce a neoplastic phenotype
similar to that seen in FAP without the need for structural alteration in the APC gene.
Indeed, occasional AAPC-like families have been described where linkage studies have
definitively excluded the APC gene as the cause of disease (Stella et al, 1993).
37 Heritable mutations in the p-catenin gene (CTNNBl) or other members of the Wnt signal transduction pathway (described below) may be able to substitute for germline APC mutations. However, to date, mutation analyses of peripheral blood DNA extracted from
'APC negative' individuals (using SSCP and direct DNA sequencing) has failed to show any hereditary alteration in the main contender, CTNNBl (Dobbie and Muller, 1999, Gao et al., 1999).
1.3.3 Somatic APC Mutations in FAP
At birth every cell in an individual with FAP, or in a Min/+ mouse, is heterozygous for mutant and WT alleles of^PC or its murine homologue Apc respectively. However, only a tiny percentage of these cells ultimately give rise to a tumour. One or more additional somatic events are necessary for neoplastic progression. It has long been recognised that the majority of tumours have mutation or loss of both/IPC alleles.
A study of 63 adenomas and carcinomas taken from a mixed population of FAP and sporadic human colorectal tumours demonstrated that more than 80% had at least one mutation in the APC gene and more than 60% showed mutation or loss of both alleles.
More than 75% of all the mutations occurred in exon 15. (Miyoshi et al., 1992b). A later study of FAP adenomas as small as 3mm in size showed that over 40% of these adenomas had mutation or loss of both APC alleles, whereas somatic K-ras mutations were rare in tumours under 1cm in diameter (Ichii et al., 1993).
Circumstantial evidence has long suggested that certain germline APC mutations have a
selective advantage over others, and that the second somatic 'hit' is not a completely random event:
38 • The 'hotspot' for germline mutations in FAP (codons 1055-1309) occurs in a small
region of the gene overlapping that where the majority of somatic mutations in both
FAP and sporadic CRC are seen (codons 1286-1503, termed the Mutation Cluster
region or MCR).
• Mutations further 5’ or 3’ of this are not only less common but also generally give rise
to an attenuated disease phenotype (see Figure 2).
Indeed, a recent study supports the hypothesis that the somatic event required to initiate tumourigenesis is pre-determined by the site of the germline APC mutation. Lamlum et al looked at >200 adenomas from 26 FAP and AAPC kindreds with mutations in the first half of the gene. Using a protein truncation test to identify APC mutation sites, over 78% of adenomas with germline APC mutations at codons 1,296 or 1,309 showed allelic loss of the wild-type allele, whilst under 4% demonstrated an exon 15 truncating mutation in the second allele. In contrast germline mutations 5’ of codon 1,194 or 3’ of codon 1,392 were rarely associated with allelic loss (under 2%) but approximately 30% of these tumours carried a second somatic mutation in exon 15 (Lamlum et al., 1999). Thus it appears that at least one mutation in the germline 'hotspot' or somatic MCR is required for tumourigenesis
Somatic allelic loss (which can arise in a number of ways) is more likely to occur as a random event than is a truncating mutation within a short specific sequence of exon 15.
Thus individuals with germline mutations within the 'hotspot' are more likely to develop a second defect in APC than those with non-classical germline mutations who generally need to acquire a somatic truncating mutation in the MCR.
In other words, mutations close to codon 1,300 have the strongest selection advantage in intestinal epithelium. Those within the MCR but away from codon 1,300 are slightly less
39 powerfiil, and the least genetically advantageous mutations are those outside the MCR.
Simple loss of function of APC is insufficient to explain this phenomenon. APC is a complex protein with multiple roles. The strongly selected region of the APC gene in colorectal neoplasia is the area known to be involved in the binding and degradation of the P-catenin protein (discussed below).
Turning to desmoid disease in FAP, germline mutations between codons 1,445 and 1,578 are strongly associated with desmoid tumours (Davies et al., 1995, Caspari et al., 1995)
(see Figure 2). Lamlum et al also demonstrated that 11 desmoids with germline mutations
5’ of codon 1,400 were associated with truncating somatic mutations of APC distal to codon 1,425. However 5 out of 6 desmoids with a more strongly selected germline APC mutation 3’ of codon 1, 449 showed WT allelic loss (Lamlum et al., 1999). This mirrors the results seen in colorectal tumours but suggests that loss or alteration of another function oîA P C may be important in desmoid disease.
1.4 The Gene in Sporadic Colorectal Tumours.
Approximately 60% of sporadic colorectal adenomas and carcinomas contain a somatic mutation of the APC gene. Over 48% of these tumours have abnormalities in both alleles of^PC (Powell et al., 1992). These mutations are similar in both type and position to the germline and somatic mutations already described in FAP, being largely small insertions, deletions or point mutations resulting in a frameshift and early truncation of the protein product. In fact, two thirds of all somatic mutations in colorectal tumours taken from a mixed population of FAP and non-FAP patients occur in a region covering only 8% of the entire coding sequence of the gene, codons 1286-1513. This has been termed the
'Mutation Cluster Region' (MCR) (Miyoshi et al., 1992b). It overlaps with the 'hotspot' for germline mutations in FAP (codons 1055-1309). (see Figure 2).
40 APC mutations occur in the smallest of sporadic adenomas (under 5mm in size), well before k-ras mutations are commonly detected (Vogelstein et al, 1988), and they are present at a constant frequency in tumours of increasing size and grade (Powell et ah,
1992). Identical mutations are also found in aberrant crypt foci (Smith et ah, 1994).
These small areas of intestinal epithelium, with irregular glandular architecture but no dysplasia, are thought to be the earliest precursors of adenomas.
1.4.1 APC Polymorphisms
Genetic polymorphism is the existence of genes in more than one form, each form being maintained by the pressures of natural selection. Two important polymorphisms have been described for the APC gene.
The APC I I 307K polymorphism (where thymidine is converted to adenosine at codon
1307) occurs only in Ashkenazi Jews at a population incidence of 5-6%. Its incidence is even greater in Ashkenazim CRC sufferers, and is as high as 28% in Ashkenazi Jewish
CRC patients with a family history of the disease (Laken et al., 1997, Rozen et al., 1999).
The polymorphism gives rise to an (A)g repeat in the MCR. Polyadenosine sequences are prone to mutation and colorectal tumours occurring in individuals with this polymorphism often have a somatic mutation of APC within the (A)g tract, generally accompanied by loss of the wild-type allele (Gryfe et al., 1998). Thus, if APC is the
'gatekeeper' to CRC (Kinzler and Vogelstein, 1996), the APCI1307K polymorphism acts as a 'susceptible gatekeeper'. Alternatively, it is possible that the polymorphism itself, with the subsequent change of amino acid residue from isoleucine to leucine, may have a direct functional effect on the APC protein. Even a subtle alteration in the binding of p- catenin, which occurs in this region, may be adequate to increase the incidence of polyp formation.
41 E1317Q is a similar APC polymorphism seen in British individuals with multiple non-
FAP colorectal adenomatous polyps (Frayling et al., 1998). This is a germline missense guanine to cytosine transversion at codon 1317. The resultant amino-acid transversion occurs in the APC domain involved in the regulation of p-catenin.
These are possibly only two of many low penetrance cancer predisposition alleles which increase the risk of cancer. The effect in an individual family may be low and of variable penetrance, but if there are multiple such polymorphisms the effect on the population could be much greater.
1.5 The APC Protein
Wild-type APC has been identified in a variety of tissues, including colon, breast, lung, lymphoid tissue, prostate and cervix (Smith et al., 1993, Bhat et al., 1994). In contrast, truncated forms of the protein are only found in ‘normal’ tissues in subjects with FAP.
The fiill length protein, like the gene encoding it, is a large molecule, made up of 2,844 amino acids with a molecular weight of approximately 310 kDa (Groden et al., 1991). It is found as part of an insoluble aggregate within the cell (Smith et al., 1993).
Mutant APC proteins have been identified in the majority of sporadic colorectal tumours, in cell lines derived from this material and in occasional tumours from elsewhere within the gastro-intestinal tract (Dihlmann et al., 1997, Chop et al., 1995, Smith et al., 1993,
Horii et al., 1992b, Horii et al., 1992a). Eighty percent of colonie tumour cell lines lack any normal APC on immunohistochemistry or immunoblotting (Dihlmann et al., 1997,
Chop et al., 1995, Smith et al., 1993). However, this must be interpreted with some caution as immunodetection of APC can be technically challenging (discussed below).
42 1.5.1 Dominant Negative and Dose Dependant Effects of The Protein
Heterodimer
The amino terminal of APC, approximately the first 900 residues, consists of proline-free heptad repeats of hydrophobic residues (apolar-X-X-apolar-X-X-X), characteristic of a helical coiled coils that allow protein-protein interaction (Groden et al., 1991, Bonneton et al., 1996) (see Figure 2).
Biochemical studies have demonstrated that amino acids^ 1-45 alone allow efficient parallel helical homodimerisation of two APC proteins (Joslyn et al., 1993). In vitro, the first 171 residues are sufficient and the first 45 amino acids are absolutely required for dimer formation (Su et al., 1993a) (See Figure 1).
A large volume of work supports the hypothesis that the APC protein also fimctions as a homodimer in vivo. Studies of CRC cell lines with varying combinations of APC alleles initially demonstrated that, in the presence of wild-type APC, mutant proteins migrate from the cytosolic compartment to join their full length partners in the membrane bound fraction of the cell (Smith et al., 1993). Subsequent work (looking at immunoprécipitation of the protein from lymphoblastoid cell lines of FAP patients heterozygous for APC), has revealed that the wild type protein and a smaller truncated form interact in vivo at the amino terminal (Su et al., 1993a).
Most APC mutations give rise to proteins over 171 amino acids in length, i.e. large enough to allow homodimerisation with their wild-type coimterpart (Nagase and
Nakamura, 1993, Beroud and Soussi, 1996, Miyoshi et al., 1992a, Miyoshi et al, 1992b,
Powell et al, 1992). This binding may inactivate or significantly decrease the activity of the wild-type protein, a so-called 'dominant negative' effect (Su et al, 1993a).
43 Germline mutations 5’ of codon 158, upstream of the homodimerisation boundary, give rise to an attenuated disease phenotype (AAPC). It has been postulated that very short
APC proteins produced as result of such mutations are unable to compete effectively for entry into the homodimer complex (Spirio et al, 1993). Indeed, researchers have consistently failed to identify intracellular APC proteins smaller than SOkD (Smith et al.,
1993). Thus there may be a reduced level of APC within the tissues of individuals with these 5’ mutations but the normal protein that is present probably retains its fimctional activity, a 'dosage' effect
1.6 Mechanism of Action of APC
The wild-type protein contains several functional domains as shown in Figure 2.
Following the aniino-terminal coiled-coil repeats (involved in protein homodimerisation) comes a short series of armadillo repeats which bind P and y catenin, and then a longer series (7 to 20) of 15 amino-acid repeats involved in the down-regulation of p and y catenin. The carboxy-terminal of the protein contains a large number of basic amino acids that bind to microtubules and other assorted proteins (discussed below).
1.6.1 p-Catenin
The major role of APC in tumourigenesis is now felt to be in its regulation of the protein p-catenin. This is a 92kD protein and a member of the 'Armadillo' family; a group of proteins with a common central domain, a minimum of 6 imperfect 42 amino acid repeats, termed 'arm' motifs. The armadillo protein of Drosophila was the first family member to be described (Riggleman et al., 1989), but a close homology to the vertebrate proteins p-catenin and y-catenin (plakoglobin) was soon noted (McCrea et al., 1991,
Franke et al., 1989).
44 In the last 10 years many proteins containing characteristic arm motifs have been cloned and sequenced (Peifer et al., 1994, Hatzfeld, 1999). Interestingly APC is also a member of this family (Kinzler et al., 1991, Groden et al., 1991). These proteins typically combine structural roles, as cell-contact and cytoskeletal-associated proteins, with signalling functions. This breadth of action is largely due to their ability to interact, through the armadillo domain, with a number of distinct partners (Hatzfeld, 1999).
1.6.1.1 The Role of p-Catenin in Cell-Cell Adhesion and Cell Motility
The adherens junction is a region of the cell membrane specialised to facilitate homotypic cell-cell interactions. It is organised around calcium dependant transmembrane glycoproteins of the cadherin family, E-cadherin being present in epithelial cells (Hirano et al., 1987).
The extracellular amino-terminal domains of cadherins on adjacent cell membranes interact directly (Blaschuk et al., 1990, Nose et al., 1990). The highly conserved cytoplasmic portions of the E-cadherin homodimer then bind specifically to the armadillo domain of p-catenin (Nagafuchi and Takeichi, 1988, Jou et al., 1995). The amino- terminal of P-catenin binds to a-catenin, which in turn mediates interaction with the actin cytoskeleton via a-actinin or through a direct interaction with actin filaments (Knudsen et al., 1995). In this way a strong cell-cell bond is formed (see Figures 3 and 4).
Nevertheless, there is a dynamic element to the cadherin-catenin-cytoskeletal complex that allows cell movement during development and wound healing. Under these conditions, epithelial migration is accompanied by tyrosine phosphorylation of p-catenin, separation of the cadherin-catenin complex and an increase in the free cytoplasmic pool o f p-catenin. (Muller et al., 1999).
45 1.6.1.2 The Role of p-Catenin In Signal Transduction and Its Interaction With APC
In 1987 Wieschaus and Riggleman identified the Drosophila armadillo gene as a key member of the 'Wingless' signal transduction pathway that decides on anterior/posterior patterns in Drosophila embryo segments (Wieschaus and Riggleman, 1987). A similar role for P-catenin in vertebrate signal transduction was suggested when the high level of homology (71%) between the proteins became apparent (Heasman et al, 1994, Funayama et al., 1995). This conservation of a signal transduction pathway during evolution suggests that it plays a central role in cellular development.
In the vertebrate signalling cascade a Wnt ligand (a cysteine rich secreted molecule associated with the extracellular matrix) binds to its receptor of the 'frizzled' family of transmembrane proteins and phosphorylates and thus activates a homologue of
'dishevelled' (Yang-Snyder et al., 1996). ('Frizzled' and 'dishevelled' were originally described in Drosophila.) This phosphoprotein in turn phosphorylates and inactivates a serine/threonine kinase, glycogen synthase kinase-3 p (GSK-3p) which is otherwise constitutively active (see Figure 3).
APC and GSK-3p proteins complex together in cells where there is an accumulation of free p-catenin. GSK-3p phosphorylates APC at conserved motifs (S-XXX-S) in the seven repeated 20 amino acid sequences (see Figure 2), allowing binding and serine phosphorylation of free cytoplasmic p-catenin by APC and GSK-3p respectively
(Rubinfeld et al., 1996, Yost et al., 1996). This alteration in phosphorylation targets the whole complex for degradation by the ubiquitin/proteasome pathway (Aberle et al.,
1997).
46 The common mutant forms of APC, lacking these phosphorylation sites and p-catenin degradation sites, can bind but not down-regulate p-catenin (Morin et al, 1997) and therefore levels of free cytoplasmic p-catenin are uncontrolled. The free p-catenin protein associates with the amino terminal of cellular transcription factors such as T cell factor
(TCP) 1 or 4, or lymphoid enhancer factor 1 (LEF 1) (Molenaar et al., 1996, Behrens et al., 1996), to form a bipartite transcription factor that enters the nucleus (Huber et al.,
1996) and promotes transcription of Wnt responsive genes such as c-myc (a potent cellular oncogene), cyclin D1 (a cell cycle regulator), PPARS, c-jun and fra-1 (all involved in transcriptional activation) (He et al., 1998, Tetsu and McCormick, 1999, He et al., 1999, Mann et al., 1999).
Thus it appears that a major function of APC lies in the regulation of intracellular levels of p-catenin and the p-catenin/TCF transcription complex. Mutations in APC have a direct effect on the Wnt signal transduction pathway and may increase transcription of a variety of cellular genes.
Wnt signalling inactivates GSK-3p and thus prevents degradation of p-catenin. In a similar manner, signalling via cell surface receptor tyrosine kinases, such as the epidermal growth factor receptor (EGF-R) leads to phosphorylation and inactivation of GSK-3P through the phosphoinositide 3-kinase (PI3-K) pathway (Eldar-Finkelman et al., 1995).
Thus signalling via receptor tyrosine kinases may both enhance the release of p-catenin from the adherens junction and stabilise the free protein within the cell.
1.6.1.3 Are the Two Roles of p-Catenin Inter-Related ?
A range of experimental data suggests that the role of P-catenin in cell-cell adhesion is distinct and separate from its role in signal transduction (Orsulic and Peifer, 1996, 47 Siegfried et al., 1994, Gumbiner, 1995). However, the final phenotype of a cell is probably influenced by the relative functional activity of each pool of protein.
1.6.1.4 The Importance of p-Catenin Alterations in Tumourigenesis
1.6.1.4.1 The Limited Abilitv of CTNNBl Mutations To Substitute For C Mutations
The frequency of APC mutations in sporadic colorectal cancer has lead many to question the aetiology of the minority of tumours (20-40%) that have two apparently normal APC alleles. In 1997, Morin et al demonstrated that 3 out of 5 colorectal tumours lacking APC mutations contained a p-catenin gene {CTNNBl) activating mutation (Morin et al., 1997).
To further clarify the relationship between mutations in APC, CTNNBl, y-catenin
{CTNNGl), GSK-3J3 and Tcf-4 genes in sporadic CRC, Sparks et al studied 24 primary human adenocarcinomas, 16 adenomas and 33 cell lines derived from adenocarcinomas
(Sparks et al., 1998). None o f 46 APC deficient tumours contained an activating mutation in CTNNBl, but 48% of 27 tumours with intact APC carried mutations in CTNNBl.
Mutations in APC or CTNNBl were mutually exclusive, and no tumours lacking mutations in these genes had abnormalities in the CTNNGl or GSK-Sp genes, suggesting that P-catenin is unique in its ability to substitute for APC mutations.
However, only a small percentage of sporadic CRCs lack APC mutations (Powell et al.,
1992). Analysis of 92 randomly selected sporadic CRCs and 57 CRC derived cell lines, demonstrated a CTNNBl mutation in only two primary tumours and one cell line, all of which also demonstrated microsatellite instability (Kitaeva et al., 1997). This suggests that although p-catenin plays a vital role in the development of the majority of colorectal tumours, mutations in CTNNBl per se are not a common initiating event in the adenoma-
48 carcinoma sequence. Rather, alterations in the expression or phosphorylation state of this oncogene are generally down stream events following another genetic insult, such as APC mutation or defective mismatch repair. Indeed, a study of 43 CRC derived cell lines known to contain an APC mutation or deletion demonstrated that all of them exhibited uncontrolled activation of the P-catenin/TCF transcription pathway (Korinek et al., 1997).
The percentage of P-catenin mutations in small sporadic adenomas (12.5%) is greater than that in large adenomas (2.4%) or invasive cancers (1.4%) (Samowitz et al., 1999).
When compared to the stable level of APC mutations seen in tumours as they progress along the adenoma-carcinoma sequence (Powell et al., 1992), this suggests that an APC mutation actually has a greater selective advantage than a mutation in CTNNBl, probably due to other tumour suppressor functions encoded by the APC gene.
1.6.1.4.2 The Effect Of Qualitative Alterations In p-catenin on Tumour Adhesion
In vitro studies of signalling through the EGF-R, hepatocyte growth factor receptor
(HGF-R), and c-erbE-2 receptor (Her-2/neu) have demonstrated immediate tyrosine phosphorylation of p-catenin (Hoschuetzky et al., 1994, Shibamoto et al., 1994, Kanai et al., 1995). Increased signalling via all of these receptors is known to occur in epithelial tumours in vivo. Activation of the tyrosine kinase ppbO^'^"^^^^ is also a frequent early event in colorectal carcinogenesis (Cartwright et al., 1990), and in vitro transformation of
Madin-Darby canine kidney (MDCK) epithelial cells with the oncogene w-src leads to tyrosine phosphorylation of p-catenin, reduced cellular adhesion, epithelial dedifferentiation and enhanced cellular invasion (Behrens et al., 1993). Increased levels of tyrosine phosphorylated p-catenin have also been seen in invasive breast cancer cell lines when compared to weakly invasive lines (Sommers et al., 1994).
49 In vivo, up-regulation of tyrosine phosphorylation of p-catenin has been observed in primary human colorectal cancers when compared to normal mucosa (Takayama et al,
1998).
E-cadherin and APC compete for p-catenin binding (see Figure 4). Their binding is mutually exclusive (Hulsken et al., 1994). Thus increased expression of APC at a cell membrane (see below) may indirectly affect cell-cell adhesion, and increase cell motility.
Alternatively, tyrosine phosphorylation of p-catenin could initiate a decrease in E- cadherin mediated cell adhesion, allowing increased migration, and secondarily providing increased substrate for APC to bind and down-regulate. The details of this important area in tumour biology have yet to be clarified.
Co-localisation of APC and p-catenin at actively migrating epithelial cell membranes has been observed on immunofluorescence, but only in a subgroup of APC protein clusters, supporting a transient rather than permanent association between the two proteins at this site (Nathke et al., 1996).
1.6.1.4.3 The Effect of Quantitative Alterations In p-Catenin on Tumour Stage and
Grade.
As cellular accumulation of P-catenin is a common initiating event in what could prove to be a plethora of downstream pathways, it has been postulated that the cellular level and/or distribution of P-catenin should correlate with tumour stage and/or grade.
Takayama et al showed a reduction in staining for E-cadherin, p-catenin and a-catenin in primary colorectal tumours. The reduction of p-catenin expression was significantly
50 correlated with dedifferentiation, Dukes stage, lymph node metastasis and liver metastasis
(Takayama et al, 1998). In contrast, Ghadimi et al found no significant reduction in expression of p-catenin in his series of sporadic CRC (Ghadimi et al., 1999).
Perhaps these apparently disparate findings are not altogether surprising. A decrease in cellular levels of p-catenin within a developing neoplasm may give rise to a less adhesive tumour with greater potential for metastatic spread. In contrast tumour initiation per se may result fi*om an elevation in p-catenin levels as a primary event or, more commonly, secondary to loss or mutation of APC. It has been demonstrated that loss of membranous
P-catenin expression, and widespread nuclear expression of p-catenin correlate with increasing dysplasia in adenomas and decreased survival following diagnosis and surgery for colorectal carcinoma (Hao et al., 1997, Hugh et al., 1999).
1.6.2 Other Postulated Functions for APC
1.6.2.1 APC and Cell Cycle Progression
Overexpression of the normal APC protein in NIH 3T3 cells slows progression of the cell cycle fi*om GO/Gl to the S phase. Overexpression of a truncated iso form of the protein does not have such a strong inhibitory effect on cell cycling, and decreases the effect of the wild-type protein when the two isoforms are administered together. This cell-cycle inhibitory effect of APC is abolished by simultaneous overexpression of cyclin E/cyclin dependant kinase (CDK) 2 or cyclin D l/ CDK 4 (Baeg et al., 1995), suggesting that APC has a direct effect on the activity of these cyclin-CDK complexes.
The APC/p-catenin/TCF pathway may also mediate this effect on cell cycling. P- catenin/TCF complexes directly activate transcription of cyclin D l (Tetsu and
McCormick, 1999). This cyclin is known to be over-expressed in many colon carcinomas, 51 and hastens progression through the cell cycle (Arber et al, 1996, Quelle et al, 1993).
Thus WT APC suppresses cyclin Dl transcription and provides a 'break' to cell cycling.
1.6.2.2 APC and Crypt Fission
In Min/+ mice, morphologically normal intestine demonstrates elevated levels of crypt fission and greater asymmetry of crypt division than does the normal intestine of control mice. The same microscopic changes are present in the grossly normal intestine of FA? patients when compared to a control group (Wasan et al, 1998).
Crypt fission is a spatio-temporal duplication process, unique to the intestinal tract. There are close homologies to axis duplication in embryonic cells, where the central role of P- catenin has been well-documented (Heasman et al, 1994). Thus changes in crypt fission, and a resulting strong intestinal phenotype, may be the most obvious result of an imbalance in the APC/p-catenin system throughout the body.
1.6.2.3 APC and Cell Migration
Mutations in APC lead to the accumulation of enterocytes near the crypt-villus transition zone in the intestine (Polakis, 1995). At a biochemical level, the carboxy-terminal residues of APC (amino acids 2130-2843), associate with cytoplasmic microtubules and promote the assembly of microtubule arrays in vitro (Munemitsu et al, 1994). In vivo, endogenous wild-type APC has also been identified at the ends of microtubules protruding into actively migrating membrane structures, suggesting that APC is involved in directed cell migration (Nathke et a l, 1996).
However, despite these interesting observations, amino-acid substitutions in the carboxy terminus of the protein are seldom seen in clinical cases of FAP or sporadic CRC.
52 There is recent evidence to suggest that APC exerts an effect on cell migration via another route. The cell surface receptor CD44 is an important mediator of cell- extracellular matrix interactions (Aruffo et al., 1990). P-catenin/TCF complexes upregulate its transcription (Wielenga et al., 1999), potentially leading to enhanced cell motility, tumour progression and métastasés (Wielenga et al., 1993). Thus WT APC may act to control the expression of CD44.
1.6.2.4 APC and DLG/EBl
The final residues of APC (amino acids 2560-2843 and 2771-2843) bind EBl and DLG proteins respectively (see Figure 2) (Matsumine et al., 1996, Su et al., 1995). EBl appears to be a component of the microtubule cytoskeleton although its function is unclear.
(Berrueta et al., 1998, Juwana et al., 1999). DLG is the human homologue of Drosophila discs large tumour suppressor protein, a guanylate kinase similar to those seen in non receptor protein tyrosine kinases and signal transduction proteins (Woods and Bryant,
1991).
However, extreme 3' mutations or truncations of^PC rarely give rise to clinical disease
(Friedl et al., 1996, van der Luijt et al., 1996, Smits et al., 1999). Mutation or loss of heterozygosity at the E B l locus is rarely seen in colorectal tumours (Jais et al., 1998), and a recently described mouse model of FAP lacking EBl and DLG binding sites but retaining some of the p-catenin binding sites does not develop polyps (Smits et al., 1999).
1.6.2.5 APC and Apoptosis
Overexpression of WT APC in the CRC cell line HT-29 leads to cell death through apoptosis (Morin et al., 1996). Transfection of a similar CRC cell line with a recombinant adenovirus (Ad-CBR) constitutively expressing just the central third of APC (the region
53 of the p-catenin binding repeats) has also been shown to prevent nuclear translocation of
P-catenin, inhibit P-catenin/TCF mediated transactivation of downstream targets, and cause significant growth arrest and cellular apoptosis. Thus it appears that the p-catenin- binding domain of APC is sufficient for its tumor suppressor activity in vitro, and that apoptosis may be an important mediator of this action (Shih et al., 2000). However, in other studies, full length APC has been expressed in SW480 CRC cells (containing a single endogenous mutant allele of the gene) with no detrimental effect on cell growth
(Hargest and Williamson, 1995, Munemitsu et al., 1995).
In vivo, the histologically normal epithelium of Min/+ animals shows an increased level of p-catenin in enterocytes, and a decreased rate of enterocyte proliferation and apoptosis when compared to WT httermates (Mahmoud et al., 1999a). Following treatment with non-steroidal anti-inflammatory drugs (NSAIDS), the tumour load in these Min/+ mice decreases, cellular expression of p-catenin drops and both proliferative and apoptotic indices return towards normal (Mahmoud et al., 1998b, Mahmoud et al., 1998a,
Mahmoud et al., 1999a).
1.7 The Interaction of NSAIDS and APC
NSAIDS are known to decrease polyp load in individuals with FAP, and in animal models of CRC (Giardiello et al., 1993, Bames and Lee, 1998, Jacoby et al., 1996,
Boolbol et al, 1996). They have also been shown to have an epidemiological effect on the incidence of sporadic CRC (DuBois and Smalley, 1996). These effects have largely been attributed to inhibition of cyclooxygenase 2 (COX-2) and a resulting decrease in cellular production of prostaglandins (Reddy et al, 1996, Tsujii et al, 1998). This is supported by the demonstration of decreased tumour burden in knockout mice who also carry a COX-2 knockout (Oshima et al, 1996), and the observation that human CRC shows 54 elevated levels of COX-2 when compared to adjacent normal mucosa (Eberhart et al,
1994, Sano et al., 1995).
However, it now appears that NSAIDS have anti-neoplastic effects that are independent of COX-2 activity (Piazza et al., 1997a, Piazza et al., 1997b). For example, NSAIDS induce change in proliferation, cell cycle phase distribution, and apoptosis in CRC cell lines devoid of COX activity, and restoration of COX induced prostaglandins into these cell lines does not restore the original state (Hanif et al., 1996). Indeed NSAIDS may act in many ways, and at all levels of the APC/p-catenin/TCF signalling pathway (see Figure
5).
Sulindac has been shown to increase the expression of APC mRNA in malignant colonic epithelial cells in vitro (Schnitzler et al., 1996). Aspirin and sulindac act in the pre neoplastic intestinal tissue of Min/+ mice to decrease an abnormally elevated level of enterocyte p-catenin back towards WT values, and to restore a normal rate of enterocyte crypt-villus migration (Mahmoud et al, 1997, Mahmoud et al, 1998b). The mechanism behind this decrease in p-catenin has not been elucidated.
In addition, the nuclear receptor PPARô (peroxisome proliferator-activated receptor ô), a ligand-dependant sequence specific transcription activator, is a common target for both
APC and NSAIDS. The PPARô promoter contains two TCF-4 binding sites. In the presence of uncontrolled signal transduction through the APC/p-catenin/TCF pathway, levels of PPARÔ increase. PPARô then stimulates transcription of other cellular genes, as yet unknown. NSAIDS act both to repress the PPARô promoter and to inhibit the binding of PPARÔ to its DNA targets. This appears to be the mechanism of action of sulindac sulphide-induced apoptosis of CRC cell-lines (He et al, 1999). 55 Therefore, it appears that NSAIDS act synergistically with WT APC to maintain suppression of transcription of certain genes that would otherwise prevent apoptosis.
1.8 Animal Models Of Colorectal Cancer
1.8.1 The Min/+ Mouse
The Min/+ mouse is without doubt the most extensively studied animal model of polyp formation in FAP and sporadic CRC (Hull et al., 1999, Wasan et al., 1997, Fichera et al.,
1997). It was first derived in a mutagenesis project at the University of Winsconsin in the late 1980s (Moser et al., 1990). C57BL/6J (B6) mice treated with a chemical mutagen, ethylnitrosourea, were seen to develop a progressive anaemia, due to the presence of intestinal polyps. The mutant gene was called Min (Multiple Intestinal Neoplasia). It was transmitted in an autosomal dominant, fully penetrant, fashion when affected male mice were crossed with B6 females to give a heterozygous Min/+ mouse. However, homozygous Min/Min progeny, produced by the intercross of two affected animals, failed to gastrulate and died in utero. Tumours were present throughout the intestine by 2 to 3 months of age, the total polyp load being 30-40, with the greatest concentration of polyps in the distal ileum.
It was soon shown that the Min locus mapped to the Ape gene, the murine homologue of
APC. The murine and human coding sequences are 86 and 90% identical at the nucleotide and amino acid levels respectively. The Min mutation is a non-sense germline mutation at codon 850, converting a leucine codon (TTG) to a stop codon (TAG) and causing premature truncation of the Ape protein product (Su et al., 1992). This mutant protein lacks the p-catenin binding site which stretches from codon 1014 to 1201 (Su et al.,
1993b). It is very similar to APC mutations documented in human colorectal tumours
(Nishisho et al., 1991, Groden et al., 1991).
56 As an experimental system Min/+ mice have several advantages over human subjects.
Each animal develops multiple polyps with an identical germline defect and they are raised under ’standard' conditions eliminating uncontrolled variation between individuals.
As a result, they have been widely used to study genetic and environmental factors affecting polyp progression.
Mom-1 {Modifier of Min-1\ on the distal arm of mouse chromosome 4, strongly modifies tumour load (Dietrich et al., 1993). The homologous site for this gene within the human genome is chromosome lp35-36, a region that displays frequent somatic loss of heterozygosity in a variety of human tumours including colonic neoplasia (Leister et al.,
1990). The gene for secreted phospholipase 2A {Pla2g2a) has now been mapped to the
Mom-1 locus and shown to display 100% concordance between allele type and tumour susceptibility in mice (MacPhee et al., 1995, Cormier et al., 1997). Its mechanism of action is not clear, although it appears to act at a cellular or crypt restricted level with no obvious systemic effect (Novelli et al., 1999). However, allelic variation in Mom-1 probably explains under half of the genetic variance in tumour load with the same APC mutation. It is likely that other modifier genes, with weaker but cumulative effects, have yet to be identified.
Pure bred colonies of Min/+ mice, such as the one used in this project, now show significantly less polyps than those initially described by Amy Moser, and survive for longer (Wasan et al., 1997), suggesting that environmental factors are also important in tumour progression. Investigation has revealed a significant increase in Min/+ tumour load and reduced survival in response to increased dietary fat content (Wasan et al.,
1997). This reflects the epidemiological trend of increasing human CRC incidence in line with animal fat content in the diet (Willett et al., 1990).
57 Over ten years ago it was noted that there is significant clustering of duodenal adenomas around the ampulla of Vater in FAP and there is a correlation between the level of duodenal polyp dysplasia and mucosal exposure to bile (Spigelman et al., 1989). A recent experiment exposed Min/+ mice to increased dietary levels of the unconjugated bile acid chenodeoxycholate and demonstrated a subsequent increase in duodenal tumour load along with increased P-catenin expression in duodenal tissue. This strongly supports the role of unconjugated bile acids as tumour promoters in the presence of an initiating Ape mutation (Mahmoud et al., 1999b). It may in part explain the greater incidence of Min/+ polyps within the small bowel (20-30 tumours) than the colon (1-4) (Wasan et al., 1997).
The vast majority of adenomas in Min/+ mice, like those in FAP, demonstrate mutation or loss of both Ape alleles. Somatic allelic loss is generally secondary to complete loss of murine chromosome 18 during mitosis (Levy et al., 1994, Luongo et al., 1994). However,
it is possible that the concomitant loss of other oncogenes and tumour suppressor genes also located on chromosome 18, such as mutated in colorectal cancer (MCC) or deleted in
colorectal cancer (DCC) respectively, are just as important in tumour initiation.
To address this issue further, a conditionally immortal colonic epithelial cell line
heterozygous for the Apc^^" mutation (IMCE) was infected with a replication defective
ras virus, T^i-v-Ha-m^. The IMCE-Ras cells were immediately transformed and rendered
tumourigenic without loss of the WT Ape allele (D'Abaco et al., 1996). Thus synergy
between a single defective Ape allele and an activated ras oncogene can initiate tumour
growth although neither mutation alone is sufficient. These two mutations are often co
existent in colorectal tumours (Vogelstein et al., 1988).
58 In vivo Min/+ mice treated with A^-ethyl-A^-nitrosurea show a significant increase in tumour burden (Shoemaker et al, 1995), and a proportion of these additional adenomas retain the WT somatic Ape allele (Shoemaker et al, 1997). Activation of the ras oncogene is not common in Min/+ mouse adenomas but the second insult may not need to be confined to ras. Thus it appears that mutation or loss of both Ape alleles is not absolutely required for tumour initiation.
In a similar vein, Bertagnolli's group have demonstrated that histologically 'normal' Min/+
small bowel epithelium (heterozygous for Ape) already exhibits elevated levels of p-
catenin on immunohistochemical analysis and that this is associated with decreased levels
of enterocyte proliferation, apoptosis and crypt-villus migration (Mahmoud et al., 1997).
1.8.1.1 Use of The Min/+ Mouse in Studies of Gene Transfer To Date
Work has been carried out over the last ten years looking at transfer of the full length
human yf PC gene into the rectal epithelium of rodents using lipofection (Westbrook et al.,
1994, Arenas et al., 1996, Fichera et al., 1997, Hargest et al., 1998). These studies have
demonstrated expression of the transgene at the mRNA level in the rectal epithelium of a
number of treated animals. However, Westbrook's group in Chicago were unable to
demonstrate rectal expression of the transgene for greater than four days after a single
treatment. When they proceeded to gene transfection in the Min/+ mouse, quantitative
PGR was used to show that only 3% of the DNA extracted fi*om treated epithelium was
derived fi-om exogenous APC after a single treatment, although six weeks of repeated
therapy increased this figure to 10% (Fichera et al., 1997). Subsequent work considered
the clinical effect of this same treatment using a protocol where four week old Min/+
mice were repeatedly treated with rectal 'gene enemas' every 72 hours for two months
prior to euthanasia. Once again the expression of the transgene was only detected up to 72
59 hours following the final treatment and there was no evidence of any alteration in disease status (as determined by the number of intestinal adenomas) (Arenas et al., 1996).
In contrast, work within our group has demonstrated that transfection of this same gene and vector at higher concentrations into a cell line containing only a single mutant copy o f the APC gene produces high level expression of the transgene with relative suppression of the endogenous message (Hargest and Williamson, 1995). Moving on to look at gene expression within the colonic epithelium of WT mice a large dose of plasmid (70pg) was administered as a gene enema in a large volume of lipofection solution (0.5ml). The transgene was expressed in up to two thirds of animals with no obvious detrimental effects and encouragingly it was detected into the fourth week after treatment (Hargest et al., 1998). Intestinal epithelium is continually turning over, thus even those epithelial cells located in the deepest part of the crypt at the time of treatment should be shed into the lumen of the bowel within seven to nine days. The continued expression of the transgene as mRNA more than ten days beyond this point suggests that a group of intestinal stem cells may have been transfected, possibly due to expansion of the crypts with a larger volume of enema and exposure to a higher dose of plasmid (Sandberg et al., 1994,
Hargest et al., 1998).
1.8.2 Other Mouse Models of FAP
The yf/7cl638N mouse carries a targeted germline mutation at codon 1638 o f the Ape gene
(Fodde et al., 1994). The mutant protein, expected to be 185kD in size, cannot be identified by immunoblotting, and is thought to be unstable. These mice develop a small number of upper GI polyps and large numbers of extra-abdominal desmoids and cutaneous cysts. They are therefore a good model of AAPC and extra-intestinal disease
(Smits et al., 1998). The attenuated intestinal phenotype and increased survival also
60 makes v4/?cl638N a good model for prolonged interventional studies and to identify genetic and environmental modifiers in tumour development (van der Houven van Oordt et al., 1999, Williamson et a l, 1999).
In order to produce a mouse model of FAP with colonic polyps, a Japanese group has developed a mouse carrying enzyme specific cutting and recombination sites (loxP) on either side of exon 14 of Ape. The mice are normal until the colorectal region is infected with an adenovirus carrying the enzyme that acts at the loxP sites. Upon viral infection and enzyme expression exon 14 is deleted and the mice develop colorectal adenomas within weeks (Shibata et al., 1997). Further work in this interesting model is awaited with interest.
In 1995 Oshima et al produced the a mouse with an Ape germline truncating mutation at codon 716 (Oshima et al., 1995). Heterozygotes {Ape^^^^^'') develop multiple polyps throughout the intestinal tract, mostly in the small intestine. With the publication of studies showing a decrease in polyp load in Min/+ animals in response to NSAIDS
(Beazer-Barclay et al., 1996, Jacoby et al., 1996) this Japanese group studied the effect of crossing Ape^^^^ mice with mice lacking one or both functional alleles of the COX-2 gene
{Ptgs2). Ape^^^^^'' Ptgs2^'' and Ape^^^^'^'' Ptgs2''' mice showed statistically significant decreasing numbers of intestinal polyps when compared to Ape^'^^^^'' Ptgs2'^'^ Httermates
(Oshima et al., 1996). This demonstrated an unequivocal role for COX-2 in the polyp progression.
1.8.3 Mouse Models of FINPCC
MSH2 is an MMR gene commonly mutated in HNPCC (Fishel et al., 1993, Liu et al.,
1994). Mice that are homozygous null for this gene {Msh2''') have been produced. These 61 mice develop lymphoma (in 80%) and intestinal tumours (in 70%) within one year. The
intestinal tumours are associated with inactivation of Ape (Reitmair et al, 1996b). In
order to examine the association between MMR gene defects and Ape mutations, male
Min/+ mice {Ape^'') were crossed with female Msh2''' mice. Ape^^' Msh2'^' mice
developed a significantly greater number of adenomas than the other genotypes. The tumours were also larger and developed more rapidly, and the mice had a decreased life
span. In contrast to the tumours normally seen in Min/+ mice, these rarely demonstrated
loss of heterozygosity o f Ape, suggesting that somatic mutation rather than allelic loss of
WT Ape is the second ’hit' (Reitmair et al., 1996a). This fits in well with the hypothesis
that mismatch repair gene defects can initiate intestinal tumourigenesis through the
introduction of replication errors into critical genes such as Ape. Further study of the
Msh2''' and Ape^^' Msh2'^' mice may elucidate environmental and genetic factors affecting
HNPCC.
Numerous other mouse models of CRC have been described, including those with
specific mutations in one or more of the other MMR genes (de Wind et al., 1999).
1.9 Gene Therapy
In its simplest form, gene therapy is any situation where nucleic acid is used as a
pharmaceutical reagant. It is generally taken to mean the placement and expression of
genes into tissues as a means of treatment or prevention of disease. Gene therapy can be
used as an isolated treatment, but is increasingly proposed in conjunction with other
therapeutic modalities (such as chemotherapy and/or radiotherapy) in an attempt to
enhance their effectiveness (Fujiwara et al., 1994b, Gjerset et al, 1995, Licht et al,
1995). This project is concerned with the use of gene therapy in the treatment and
prophylaxis of CRC.
62 1.9.1 The History of Human Gene Therapy
Over the last 15 years the annual published Medline references to gene therapy have risen
exponentially, from approximately 20 in 1985, to 200 in 1990 and 2000 in 1999. This reflects advances in the fields of lipid production, viral manipulation, viral replication,
and gene cloning. Indeed, the human genome project has recently published a first
working draft of the entire human genome sequence.
The potential beneficial effects of gene transfer were apparent over 20 years ago, when it
was shown that cloned genes (such as p-globin, which is mutated in thalassaemia) could
be inserted into cell lines and transcribed and translated to produce a ftmctional protein
not normally expressed by the cell (Mulligan et al., 1979). Mercola and Cline soon went
on to show that the herpes simplex virus thymidine kinase gene {HSV tk) could be
introduced into the bone marrow cells of mice ex-vivo and that the viral sequence was
expressed in vivo when the cells were re-introduced into the irradiated donors (Mercola et
al., 1980, Cline et al., 1980). Following on from this experiment, in 1980, Cline and
colleagues attempted to transfect bone marrow cells of thalassaemic patients with a
'normal' human P-globin gene ex vivo, and re-inftise them into the host, in an attempt to
restore P-globin function. Unfortunately the study failed and the whole issue of human
gene therapy and its control came into question (Wade, 1980, Wade, 1981). As a result of
the subsequent investigation, the Recombinant DNA Advisory Committee (RAC) and the
Gene Therapy Advisory Committee (GTAC) were set up in the United States and Great
Britain respectively in order to control all human gene therapy trials in these countries.
The first federally approved trial of human somatic gene therapy was finally carried out in
the United States ten years later. Severe combine immunodeficiency (SCID) is a
potentially fatal disease caused by germline mutation of the enzyme adenosine deaminase
63 (ADA). It is characterised by the presence of non-functional T-cells and recurrent
overwhelming infections. In 1990, two children with SCID were treated by infusion of
autologous T cells transfected ex-vivo with the 'normal’ ADA gene, using retroviral mediated gene transfer. This trial was successful both in terms of clinical response and T-
cell function (Blaese et al., 1995).
An escalating number of human gene therapy trials are now in progress. The majority are
in cancer gene therapy as these patients have a terminal disease, often fail to respond to
conventional therapy and are suitable for phase I and II trials of toxicity and therapeutic
efficacy.
1.9.2 Gene Delivery
At the moment, the greatest barrier to effective gene therapy is the absence of an ideal
gene delivery system. The perfect system requires a once only treatment with specific and
complete targeting of all abnormal tissue and a high efficiency of exogenous gene
transduction. It should be a safe treatment, with minimal side effects, and one that induces
a minimal host immune response. The vehicle used to introduce a gene of interest into the
target tissue is called a vector. There are many potential vectors. They are most easily
considered as viral and non-viral vectors.
1.9.2.1 Viral Vectors
1.9.2.1.1 Retrovirus
Replication incompetent retroviruses are commonly used as vectors in animal and human
trials of cancer gene therapy (Ram et al., 1997, Roth et al., 1996, Devereux et al., 1998).
The viral genes necessary for replication are deleted and replaced with the therapeutic
gene, along with its 5' promoter and controlling sequences. 64 Retroviruses target actively dividing cells. Although this makes them unsuitable for gene transfer into stable tissues, it is usefiil in targeting tumours. The virus is also able to
integrate stably into the host chromosomal DNA and could thus be expected to exert its
effect not only in transfected cells but also in all daughter cells. However, retroviruses are
less efficient transfection vectors than adenoviral constructs, and safety considerations
remain a major concern (Vile and Russell, 1995).
The random nature of retroviral integration in the host genome raises the possibility of
insertional mutagenesis. However, over 1600 patients have been treated in clinical trials
of retroviral gene therapy to date without evidence of a single recombination event
(Schagen et al., 2000). There is also a theoretical risk of recombination between the
vector and helper viruses (present in packaging cell lines) to produce a replication-
competent, pathogenic virus.
1.9.2.1.2 Adenovirus
Adenovirus serotypes 5 or 2, and recombinant derivatives thereof, are probably the most
popular vectors for gene therapy. They are trophic for a wide variety of mammalian cell
types (Kremer and Perricaudet, 1995) and can enter cells at any stage of the cell cycle.
Once internalised by endocytosis, the adenovirus is able to disrupt the endosome and
evade destruction. Thus it is highly efficient at cell transduction. Adenovirus is also
produced at a high titre by packaging cell lines (10*^ compared to approximately 10^ for
retrovirus producing cell lines). Despite these advantages, adenoviral vectors, like
retroviruses, are dogged by problems.
First or second generation replication defective viruses (where early viral genes are
deleted and replaced with novel genetic information) are traditionally used in gene 65 therapy. However, recombination events between the deleted adenoviral genes and complementary sequences integrated into packaging cell lines can produce replication- competent, pathogenic organisms (Lochmuller et al., 1994). In addition, externally introduced replication competent adenoviruses (RCAs) can contaminate the production line. RCAs replicate in an uncontrolled fashion in immunocompromised individuals, and generate an inflammatory response in immunocompetent patients. Screening for RCAs is therefore mandatory prior to clinical use (Dion et al., 1996), and has increased the cost and time required for the production of safe adenoviral vectors.
Even in the absence of recombination events, the immune response of the host to an adenoviral vector plays a central role in clinical outcome. Although in one respect the
immune response to the vector may enhance any bystander effect (see section 1.9.3.1.1), it also potentially hastens viral loss and it may induce significant inflammation in normal tissue. This makes repeated dosing with the same viral serotype dangerous. Unfortunate
individuals have even developed overwhelming adverse reactions following a single
treatment, probably due to pre-existing immunity acquired through an earlier respiratory
infection (Lehrman, 1999, Chirmule et al., 1999).
At a practical level, the size of exogenous DNA insert is limited to approximately 4-5Kb
in conventional adenoviral vectors. This can present a problem when complete genes and
their controlling sequences need to be transfected. In addition, adenoviral vectors do not
integrate into the host DNA (Schagen et al., 2000), so the duration of transgene
expression is limited to the time taken to clear the virus.
66 1.9.2.1.3 Adeno-associated virus TAAV)
AAV is a small (5Kb) replication defective parvovirus in which the majority of the
intrinsic viral sequence is removed and replaced with a gene of interest. This decreases the antigenic nature of the vector but the amount of DNA that the virus can transfer is
limited. In the presence of a helper virus (adeno or herpes virus), AAV enters a wide range of mammalian cells, regardless of mitotic state. Unfortunately the requirement for a
helper virus increases the risk of recombination events and reduces the viral titre (to
approximately 10"^).
AAV has been used to administer oral gene therapy, as it appears capable of entering
intestinal epithelium (Rabinowitz and Samulski, 1998) and resists extremes of temperature and pH. Indeed, persistent expression of p-galactosidase was detected in the
lamina propria and in the intestinal brush border of enterocytes for upto 4 months after
oral administration the gene encoding it within an AAV vector. The vector itself was
present for 6 months (During et al., 1998).
1.9.2.1.4 Other Viral Vectors
HSV, Sendai, vaccinia, polio, lentiviruses and other viruses have been investigated with
respect to their potential role as gene delivery systems. They are not the forerunners in
laboratory based or clinical trials.
1.9.2.2 Non-Viral Vectors
Non-viral gene delivery offers to eliminate the problems of viral immunogenicity and
reversion to the infectious phenotype, at the same time providing a cheap, easy to use,
plentiful supply of vector.
67 Naked DNA, in plasmid form, has been successfully introduced into muscle, skin and
brain in vivo with a good clinical response, in the form of antibody production or
transgene expression (Wolff et al., 1990, Hengge et a l, 1996, Schwartz et al., 1996).
However, the efficiency of gene transfer is not high and other non-viral vectors are under
investigation.
1.9.2.2.1 Lipofection
Liposomes contain cationic lipids that coalesce with negatively charged plasmid DNA to
form a complex. Additional neutral lipids are also present to enhance uptake into cells by
endocytosis. Thus genes of interest can be delivered into a variety of cell types, uptake
being only dependant upon cell surface exposure to the complex, and not limited by the
size of DNA carried. The majority of endocytotic vesicles fuse with lysosomes, leading to
lysosomal destruction of the plasmid. However, some vesicles escape this fate and deliver
their contents safely to the cell cytoplasm.
The transient gene expression seen with lipofection will undoubtedly require repeated
treatments to produce a long-term effect. The safety of liposomes has been demonstrated
over many years in both cosmetic and pharmaceutical preparations. For example,
AmBisome® is liposomal amphotericin, widely used in the treatment of systemic fungal
infections. However, evidence is still accumulating about repeated use internally, but as
yet no significant problems have been identified.
Although the in vitro efficiency of lipofection is good (Feigner et al., 1987), early in vivo
studies of intravenous administration of liposome/DNA complexes failed to give the same
results (Zhu et al., 1993, Liu et al., 1995b, Aksentijevich et al., 1996). This was probably
due to serum-induced disintegration of the lipid vector and subsequent DNA release and 68 degradation (Li et al., 1999). In recent years the use of more resistant combinations of neutral and cationic lipids has increased the efficacy of this system (Templeton et al.,
1997). The addition of ligands to the positively charged outer envelope of the lipid/DNA complex has also been shown to increase specific tissue targeting after systemic
administration (Templeton et al., 1997).
As with viral vectors, the route of administration of liposome mediated gene therapy
affects the level and site of transgene expression. Direct application of liposome/DNA
complexes to an epithelial surface, such as the rectum or the lungs, appears to increase
transfection efficiency to almost 100% (Westbrook et al., 1994, Hargest et al., 1998,
Stribling et al., 1992). Intra-peritoneal delivery of liposome /DNA complexes has been
used in animal models of peritoneal seeding fi-om carcinoma and glioblastoma with
effective local transfection of tumour cells (Aoki et al., 1997, Kikuchi et al., 1999, Hsiao
et al., 1997). Normal tissues were largely unaffected, probably due to a decreased level of
cell proliferation relative to the tumour. Oral administration of liposome mediated gene
therapy has not yet been extensively studied. This project covers rectal, gavage and intra-
peritoneal routes of administration of the APC gene in a liposomal vector.
1.9.3 Cancer Gene Therapy
Although gene therapy was initially envisaged as a treatment for monogenic disorders,
such as cystic fibrosis or SCID, it is now most often considered in the management of
cancer and other diseases that arise from a combination of genetic and environmental
insults. Corrective gene therapy is only one approach to treatment. In fact, at present,
gene therapy is more widely used to optimise existing cytotoxic or immunological
treatments.
69 1.9.3.1 Cytotoxic Gene Therapy
1.9.3.1.1 Gene Directed Enzyme Prodrug Therapy TGDEPT)
A gene, whose protein product is capable of activating a non-toxic prodrug to an active cytotoxic compound, is introduced into tumour cells. The patient is subsequently treated with the appropriate prodrug and transfected tumour cells are preferentially killed. Non transfected tissue cannot activate the prodrug and remains largely unharmed. The enzyme prodrug combinations most widely used are HSV tk with ganciclovir and cytosine deaminase with 5-fluorocytosine. Interestingly, although only a fraction of targeted cells
are transfected, the treatment enhances cell death in a greater overall percentage of cells.
This so called 'bystander' effect is presumably due to movement of toxic metabolites into
nearby cells (Bi et al., 1993) or effects on the local vasculature (Ram et al., 1994).
Brain tumours consist of actively dividing cells on a background of non-replicating tissue
and they are often resistant to conventional treatments. Therefore, they were the first
tumours to be treated with retroviral mediated GDEPT. Rat gliomas were injected with
retrovirus carrying HSV tk and the animals were subsequently treated with ganciclovir.
Significant tumour regression was seen, with no effect on surrounding neural tissue (Ram
et al., 1993). Similar protocols have been used with promising effects in Phase I and II
clinical trials for the treatment of human brain tumours (Ram et al., 1997, Valery, 2000).
Adenovirus mediated GDEPT is now being trialed in the treatment of metastatic CRC.
The adenoviral vector, carrying cytosine deaminase, is injected directly into hepatic
métastasés and the patients are subsequently treated with oral 5-fluorocytosine (Crystal et
al., 1997).
70 1.9.3.1.2 Gene Therapy To Promote The Therapeutic Index of Cytotoxic Drugs
The transfer of drug resistance genes to haemopoeitic stem cells could in theory allow the use of increased doses of myeloablative chemotherapy in cancer (Ward et al., 1994). The multidrug resistance gene, MDR-1, has been widely studied in this context. It encodes a cellular efflux pump, P-glycoprotein, which rapidly removes hydrophobic chemotherapeutic agents from cells (Pastan and Gottesman, 1991). Autologous bone marrow, or blood-derived stem cells, can be removed from a patient, transduced with
MDR-1 and re-infused prior to high dose chemotherapy with the appropriate myelotoxic drugs.
Studies of this type of gene therapy in animals have been encouraging, suggesting effective marrow resistance to chemotherapy (Hanania et al., 1995) and selective enrichment of treated cells groups (Allay et al., 1998). Phase I clinical trials are underway in humans with haematological malignancy and breast cancer (Devereux et al., 1998,
Hesdorffer et al., 1998, Moscow et al., 1999).
However, there are inherent flaws in this strategy that have yet to be countered. First, protection of haemopoeitic stem cells does nothing to reduce drug toxicity in other tissues, so the potential for dose escalation is limited. Second, but no less important, malignant cells within the bone marrow may be transduced, reducing the efficiency of chemotherapy and providing a selective growth advantage to this sub-group of neoplastic cells. Finally, a significant clinical effect has still to be demonstrated.
1.9.3.2 Immunological Gene Therapy
Immunotherapy aims to boost the size and increase the specificity of the host immune response to a tumour. This is an attractive proposition, as enhanced immune recognition
71 of the primary tumour may allow amplification of the response to attack disseminated disease. However, it is difficult to identify useful therapeutic genes as each tumour expresses different combinations of histocompatability antigens, cytokines and tumour specific antigens.
1.9.3.2.1 Non-specific Immune Enhancement
Non-specific immune enhancement is achieved through the introduction of genes
encoding cytokines and/or allogeneic histocompatability antigens into tumour cells or immune effector cells.
Loss or reduction of expression of cell surface class I major histocompatability complex
molecules (MHC) is often seen in CRC (Smith et al., 1989). This may allow the tumour to escape immune surveillance. Studies of transfer of Class I MHC into CRC hepatic
métastasés have been carried out, in an attempt to boost anti-tumour immunity. Fifteen
patients were treated in a phase I study, using a liposomal vector. Transgene expression
was obtained at the cDNA, mRNA and protein level and no serious toxicity occurred
(Rubin et al., 1997). Further studies are needed to establish therapeutic efficacy.
In another clinical trial adenoviral vectors containing interleukin-2 were injected into
unresectable CRC. The aim was to enhance local cytokine production and local immune
response to the tumour, without the effects of systemic administration of cytokines (Gilly
et al., 1998). No toxicity was seen, there was increased cell surface expression of
interleukin-2 receptors and one patient showed tumour necrosis.
The main drawback to non-specific immune enhancement is the potential for unwanted
systemic effects and the initiation of autoimmune phenomena. 72 1.9.3.2.2 Specific Immune Enhancement
Specific immune responses can be boosted with the use of genes encoding tumour- associated antigens. Carcinogenic embryonic antigen (CEA) is expressed on many colorectal cancers. A plasmid has been engineered for human administration that carries the gene for CEA under a CMV promoter. It has been shown to induce CEA-specific lymphoblastic transformation and antibody response in mice and protect them against later tumour challenge with CEA-expressing CRC cells (Conry et al., 1995).
A Phase I clinical trial of this plasmid in patients with metastatic CEA-expressing CRC, showed a good antibody response to CEA and no toxicity (Conry et al., 1997). Animal work suggests that the antibody response could be further boosted by the co administration of genes encoding other cytokines (Conry et al., 1996).
1.9.3.3 Corrective Gene Therapy
The aim of corrective gene therapy is to restore a normal genotype to a malignant or potentially malignant cell. This can be via the introduction of a tumour suppressor gene, by inactivation of an oncogene or a combination of both approaches. Restoration of a
’normal' phenotype is unlikely with single gene approaches, but corrective gene therapy
may be useful in enhancing the efficiency of tumour killing in response to conventional therapeutic agents.
1.9.3.3.1 Conceptual Problems With Corrective Gene Therapy
It has been said, "No therapy directed to a single dominant or recessive oncogene is likely to become a panacea" (Bishop, 1991). The main objection cited is that cancer is a
multistep process and correction of one genetic error is inadequate to slow, halt or reverse this process. Concern has also arisen over the present inability of gene therapy to target 73 all tumour cells, in case this leads to positive selection of a sub-group of untreated cells.
The need for lifelong repeated treatments to maintain 'normal' gene expression is also unattractive.
Although cancer is a multistep process, there are certain genes whose mutation appears critical in the initiation of the process or in development of the malignant phenotype. APC and P53 tumour suppressor genes have been identified as playing important roles in the development o f CRC. P53 is also commonly mutated in many other cancers, providing cells with resistance to apoptosis amongst other selective advantages (Steele et al., 1998).
Therefore, much experimental work has looked at the effect of replacing WT p53 into malignant cells in vitro and into tumours in vivo.
CRC cell lines containing mutations in p53 were transfected with the WT gene in a plasmid vector. Successfully transfected cell lines demonstrated decreased growth on soft agar and failure to progress through the cell cycle (Baker et al., 1990). Similar experiments transfected WT p53 into an osteosarcoma cell line lacking the endogenous
gene. Once again, expression of the WT gene suppressed the neoplastic phenotype of the
cells (Chen et al., 1990).
Non-small cell lung cancer (NSCLC) commonly contains p53 mutations. WT p53 was
introduced into orthotopic NSCLC, in a nude mouse model. Tumour formation was
suppressed in over 60% of treated mice when compared to control animals receiving only
the retroviral vector (Fujiwara et al., 1994a). Similar results have been demonstrated Avith p53 gene therapy in animal models of ovarian and pro static adenocarcinoma and
squamous cell carcinoma of the head and neck (SCCHN) (Mujoo et al., 1996, Asgari et
al., 1997, Liu et al., 1995a, daym an et al., 1995). Thus addition o f a single WT tumour
74 suppressor gene is able to specifically suppress tumour cell growth in vitro and in vivo in cells known to contain mutations in numerous other genes. This may be via a range of
effects, including improvement in apoptosis and an anti-angiogenic effect (Shao et al.,
2000, Bouvet et al., 1998a).
Corrective gene therapy with p53 also exerts a synergistic effect with chemotherapy
and/or radiotherapy in tumour cell killing. WT p53, in an adenoviral vector has been
shown to enhance apoptosis and tumour suppression in response to cisplatin in NSCLC
and CRC cell lines in vitro. Therapeutic co-operation with cisplatin and ionising radiation
has also been demonstrated in orthotopic lung cancers and glioblastomas in vivo (Nguyen
et al., 1996, Fujiwara et al., 1994b, Gjerset et al., 1995, Ogawa et al., 1997).
Although critics of corrective gene therapy express concern over its present inability to
target all affected or at risk tissue, this inadequacy is accepted without question in the use
of conventional treatments, such as chemotherapy or radiotherapy. Surely any decrease in
malignant tumour burden is an advantage, even if only for symptomatic relief. In
addition, reduction in tumour load may improve the ability of the host immune response
or conventional therapies to deal with the remaining disease. Advances in vector
technology and optimisation of the bystander effect also aim to improve the efficiency of
tumour cell kill.
The effect of many current gene therapy interventions is temporary but newer retroviral
vectors and emerging replication competent viral vectors offer the future possibility of a
single treatment with continued gene expression in the progeny of treated cells. Once only
treatment could also be achieved by stem cell targeting. Stem cells in the blood can be
identified by a surface marker (CD34). This has been utilised to extract a stem cell
75 enriched fraction for transfection with the ADA gene in SCID (Mitani et al., 1993).
Unfortunately there is as yet no means of identifying intestinal stem cells, or the putative stem cells in the majority of other organs. This is an ongoing area of research.
1.9.3.3.2 Antisense
Although most attention has focused on corrective gene therapy using WT tumour suppressor genes, methods of inactivating mutant oncogenes have also been developed.
Antisense oligodeoxynucleotides (ODNs) bind to complementary mRNA and prevent its translation. Thus specifically engineered ODNs block the expression of targeted genes
(Wagner, 1994). Initial studies suggested that ODNs were highly variable in efficacy and reliability, and associated with a variety of non-specific, cytotoxic, growth inhibitory and cell cycle effects (Galderisi et al., 1999). However, there is experimental evidence to show that they can suppress the growth of tumours in animal models (Sacco et al., 1998,
Cotter et al., 1994), and they are now being trialed in human disease.
1.9.3.3.3 Human Trials of Corrective Gene Therapv
Phase I trials of a retroviral vector containing WT p53 under a p-actin promoter have been carried out in the USA, for the treatment of NSCLC where conventional therapy failed to control disease. Nine patients were treated by direct intra-tumoural injection.
Three showed tumour regression and three showed stable disease. No significant toxic effects were noted (Roth et al., 1996). Similar results have been seen in trials of WT p53 transfer into NSCLC using an adenoviral vector (Swisher et al., 1999).
Patients with incurable SCCHN have been treated with multiple intra-tumoural injections of p53 in an adenoviral vector. No significant toxicity was noted. Out of 17 patients with
76 imresectable tumours, two showed a partial response, and six patients showed stable disease for over three months. One of sixteen patients with resectable tumour showed a complete pathologic response (dayman et al, 1998).
Human trials are currently looking at the effect of p53 gene therapy in combination with chemotherapy and/or radiotherapy in NSCLC and SCCHN.
1.9.3.3.4 FAP and Sporadic CRC Corrective Gene Therapy
CRC should be a good candidate for corrective gene therapy as it is one of the few cancers in which the underlying genetic alterations have been identified to any extent.
The majority of sporadic CRCs, and all those associated with FAP, are known to proceed along the adenoma-carcinoma sequence (Fearon and Vogelstein, 1990) (see Figure 1), so
an obvious target for corrective gene therapy in these tumours is the 'gatekeeper' gene
APC.
WT APC transfection into CRC cell lines containing only mutant APC has been shown to
cause a pronounced reduction in total p-catenin levels, by increased breakdown of excess
cytoplasmic p-catenin (Munemitsu et al, 1995). It also leads to increased cell death
through apoptosis (Morin et al, 1996). Either or both of these effects may be sufficient to
promote tumour cell death in vivo.
1.9.3.3.4.1 Prophylactic Gene Therapy
It may be possible to target gene therapy at an early stage, to prevent the development of
sporadic or FAP related colorectal adenomata, or to prevent their progression to
carcinoma. In FAP important areas for prophylactic therapy include the rectal stump after 77 IRA, the duodenum, and tissues giving rise to desmoid disease, such as small bowel mesentery. Safety issues surrounding gene therapy are always important, but they become vital in the consideration of prophylactic treatments, where the patients are ’healthy' individuals rather than people with pre-existing malignancy.
1.9.3.3.4.2 Aims of this Project
High levels of exogenous gene uptake, suppression of the endogenous mutant gene and prolonged expression of the transgene are the basic requirements for prophylactic gene therapy and have been demonstrated in Miss Hargest's work on APC gene transfer into
WT mice. This study was set up to see if similar high levels of transgene expression could be obtained in the intestinal epithelium of the Min/+ mouse, via rectal and upper gastro
intestinal administration of the transgene, and to determine whether there were any phenotypic effects resulting from gene expression.
Although FAP largely affects the colon, rectum and duodenum with relative sparing of
the remainder of the small intestine, murine disease is most severe within the small bowel
(Moser et al., 1990, Wasan et al., 1997, Arenas et al., 1996). Therefore any attempt to see
a significant alteration in an already low rectal or colonic polyp load, as a result of local
gene therapy, would require treatment of a large number of mice. This entails obvious
practical and ethical considerations. In the hope of recognising the presence or absence of
a significant effect more easily it was decided to consider additional treatment end-points.
Based on knowledge of the function of APC the surrogate end-points chosen included
expression of the protein product of the transgene and alteration in the binding and
intracellular expression of the protein P-catenin, whose degradation is controlled by APC.
Finally it was decided to investigate the feasibility of intraperitoneal gene administration
in WT mice, as an initial step in the consideration of gene therapy for desmoid disease. 78 Figure 1-The Adenoma-Carcinoma Sequence
Normal ^ MMR Deficiency
APC (or p-catenin) Loss or Mutation
Cell Proliferation
Formation of Aberrant Crypt Foci
Loss of DNA Méthylation
Early Adenoma MMR K -ras M utation Deficiency i Intermediate Adenoma
18q Loss (Includes DCC and Other Loci)
Late Adenoma
p53 Mutation/ 17p Loss
Carcinoma Loss of other tumour supressor
I genes, mutations of oncogenes,
▼ alterations in angiogenesis Métastasés
Based on Kinzler and Vogelstein’s model, Cell 1996 [Kinzler, 1996 #59]
79 Figure 2-APC Structure and Function (modified from Kinzler & Vogelstein, 1996) 0 171 900 2843
Amino acid Sequence 2 ^ 5 2 ^ 0 2560 2771 \/ Q, T T'f' V V V
Sequence Features m r n n
Heptad repeats 15-aa repeats 20-aa repeats Basic region t v X-COOH
‘arm” repeats
Homodimerisation p-Catenin binding site
Functional _
I1307K E1317Q Site of APC Polymorphisms Sporadic l / Colorectal Mutation Cluster Region ^ * Phosphorylation site 4 Cancer 1285 1554 Figure 3-Mechanism of Action of APC and P-Catenin
Receptor Tyrosine Kinase
E-Cadhcrin
adhesion caienin catenin
Wnt ligand ■v_A\>-vcytoskeleton
Mutant i
Cell motility catcnin Crypt Fission
Apoptosis
Transcriptional activation 1^ Multiple Effects
Tyrosine phosphorylation Serine/Threonine phosphorylation
In the presence of ligands for repeptor tyrosine kinase or Wnt receptors, GSK-3P is inactivated. This prevents the normal down-regulation of (3-catenin. Thus the free protein enters the nucleus and complexes with Tcf-4/LEF to activate transcription of a variety of genes. Germline or somatic mutation of the APC gene produces a truncated protein which can bind but not degrade p-catenin, giving the same final effect. Sequestration of p-eatenin in this way may decrease eell-eell adhesion at the adherens junction, with subsequent loss of ‘invasion suppressor’ function. Aside from these roles in signal transduction and cellular adhesion, APC mutation may directly effect cell migration, crypt fission and apoptosis.
81 Figure 4-P-Catenin Structure and Function
Sequence Features CD N c
Armadillo repeats
Functional Domains
a-catenin binding sites
APC binding sites
E-cadherin binding sites
TCF/LEFl binding sites
^ 7 ^ Serine/ Threonine Phosphorylation site
82 Figure 5-The Interaction of APC and NSAIDS Wild-type APC
Increased Transcription of APC
Sulindac Increased DOWN Decreased level of (3- NSAIDS REGULATES catenin ? Direct effect production Pla2g2a (Moml)
P-catenin Xrachidonic InhibitiotN^ Acid woo Tcf-4
Repression Inhibits binding COX 1 ofPPARÔ of PPARÔ to Ceramide promoter downstream targets Prostaglandins Transcription
Apoptosis / Transcription ot crc/in 1)1 unknown genes
APC and NSAIDS act to inhibit a common target, the nuclear receptor and transcription aactivator PPAR8. Chapter 2
Mouse Genotyping
84 This chapter describes how a crude extract of mouse genomic DNA (from an easily accessible tissue source) was differentially amplified by two pairs of primers in the polymerase chain reaction (PCR) to differentiate between mice which were genetically
Min/+ heterozygotes and their WT littermates.
2.1 Materials and Methods
2.1.1 DNA Extraction
2.1.1.1 Protocol
• Ear snips were taken from mice at the time of death (for initial comparative studies in
untreated animals) or at 21 days of age (prior to treatment randomisation during the
clinical trials).
• Each ear snip was placed in a labelled tube containing 50pl of ear lysis buffer
(Appendix 1) with 5pi of proteinase K and left overnight at 55°C to allow protein
digestion.
• Samples were centrifuged at 13,000g for one minute to pellet cellular debris.
• 80pl of H 2 O was added to dilute the sodium dodecyl sulphate (SDS) in the lysis
buffer. Samples were then boiled at 100°C for 20 minutes to destroy the proteinase K.
• After cooling to room temperature, samples were centrifijged again at 13,000g for one
minute and the supernatant was used as a source of DNA.
Long term storage was at -20^C, short term at 4^C.
2.5pi (approximately 50-100ng) of this DNA solution was used as the template for the
PCR reaction.
85 2.1.2 PCR Amplification
PCR is a wonderfully simple concept, first suggested thirty years ago (Kleppe et al.,
1971) and first used in 1985 (Saiki et al., 1985). It allows selected regions of the genome, or samples of nucleic acid, to be amplified in vitro by a factor of at least 10^. The main requirement in selection of a fragment for amplification is a knowledge of the nucleotide sequences flanking it. Short oligodeoxynucleotide sequences complementary to these flanking regions (called primers) can then be obtained commercially.
The whole process can be split into three phases, dénaturation, annealing and extension.
• Primers are added to the template DNA in a solution of salts and buffers. The
temperature is ramped up to over 90®C to denature the DNA being amplified. This
reversibly separates the hydrogen bonds holding the strands together to give two
complementary single strands.
• After one to two minutes the solution is cooled towards a physiological temperature.
This allows hydrogen bonds to re-form and primers, present in excess, anneal to
complementary sequences on the template. Raising the annealing temperature and
adjusting the salt concentration of the reaction mix increases the stringency with
which primers anneal.
• Once the primers have annealed, to form a short section of double stranded DNA, a
thermostable DNA polymerase catalyses the addition of individual deoxynucleotide
triphosphates (dNTPs), complementary to template DNA, onto the 3' end of the
primer. This is the extension phase. Taq polymerase from the thermophilic organism
Thermus aquaticus is available commercially. It can tolerate temperatures up to 100®C
and is thus ideal to replicate DNA under the most stringent conditions (Chien et al.,
1976).
86 The cycle is repeated 30 to 35 times, producing an exponentially increasing number of copies of the target DNA, limited only by the amount of Taq and dNTPs in the reaction mix. The nature of the reaction explains the phrase polymerase 'chain reaction'. It is generally carried out in a heating block, which can be programmed separately for each
reaction.
The genotyping PCR reaction used here is a triple primer PCR designed to differentially
amplify two products dependant upon the genotype of the animal (see Figure 6). Primer
OÏMR034, complementary to nucleotides 2841 to 2859 of the murine Ape sequence,
adheres to both Min/+ and WT alleles. Primer 0IMR033, complementary to nucleotides
2241 to 2258 adheres to WT alleles. Primer oIMR758 recognises the Ape sequence from
bases 2529 to 2549, containing the Ape mutation. Under stringent conditions this
primer does not anneal to the WT Ape allele. Thus the mutant allele is amplified to give a
300bp product and the WT allele is amplified to give a 600bp product. If both alleles are
present, as in the Min/+ animal, both products are formed. If the animal is a WT mouse
only the 600bp product is seen. All primers were obtained commercially (Sigma-Genosys
Ltd, Cambridgeshire, UK).
2.1.2.1 Protocol
• 2.7pl of template DNA was placed in each PCR reaction tube with
1.25 pi 20pM stock Primer oIMR034
0.625pl 20pM stock Primer oIMR758
0.25 pi 2pM stock Primer oIMR033
2.5pl 2mM dNTPs
2.5pl lOx Supertaq buffer
2.5 units 'Supertaq' polymerase
87 • The reaction mixture was made up to 25pi with ddH20, and covered with mineral oil
to prevent evaporation during the dénaturation phase.
• Negative control tubes, with no DNA or no ’Supertaq' polymerase, were included in
all experiments. Positive controls in the form of DNA from known Min/+ and WT
animals were also included.
• The PCR programme was run as shown (Appendix 1).
• 12.5pl of each PCR product was mixed with 2pl Orange G loading dye and loaded
onto a 2% agarose gel for analysis (described below).
2.1.3 Gel Electrophoresis of DNA
Fragments of DNA can be separated by molecular weight using electrophoresis through agarose or polyacrylamide gels. The final position of the nucleic acid is identified in relation to markers of known size. Agarose gels separate fragments of DNA over a broad
size range, from 200bp-50Kb. Increasing the concentration of agarose increases the resolving power of the gel. Polyacrylamide gels have an even higher resolving power and
are used to separate smaller fragments of DNA (5-500 bp). Due to the size of nucleic
acids being handled, agarose gels were used in this project.
2.1.3.1 Agarose Gel Electrophoresis
Agarose is a linear polymer of D-galactose-0-3,6-anhydro L-galactose. It dissolves at
high temperature in aqueous buffers, such as TAB (Appendix 2). When this solution cools
it forms a matrix, the density of which is determined by the concentration of the agarose
within it. When samples of nucleic acid are loaded into the gel and an electric field is
applied across it, the negatively charged bases migrate through the matrix towards the
anode. The rate of movement is inversely proportional to the molecular size of the DNA,
88 although it is affected by other variables such as the agarose concentration, the voltage applied and the electrophoresis buffer used.
Intercalating dyes such as ethidium bromide (EtBr) are added to the gel at the end of electrophoresis. EtBr intercalates between the base pairs and is visible under ultraviolet
(UV) illumination.
2.1.3.1.1 Preparation of Agarose Gels
• The edges of the horizontal gel plate were sealed with adhesive tape.
• A 2-3% agarose suspension in TAE was heated in an Erlenmeyer flask in the
microwave until clear. Intermittent shaking prevented superheating. 2% gels were
chosen for genotyping for ease of handling and because they give good resolution of
DNA fragments from 300-600bp in size.
• Whilst the gel solution was still hot a plastic comb was placed at one end of the gel
plate in order to produce wells within the gel into which the DNA samples could be
loaded.
• Once the gel had polymerised, it was placed in the electrophoresis tank and the
remainder of the buffer solution was added until the surface of the gel was
submerged.
2.1.3.1.2 Running Agarose Gels
• Each DNA sample was pre-mixed with a small volume (2pl) of a marker dye Orange-
G and carefully loaded into a different well. The marker dye functions to increase the
density of the sample so that it falls smoothly to the bottom of the well. Its colour both
simplifies the loading process and allows the DNA migration to be observed with the
naked eye. Ideally with a simple mixture of DNA 100-500ng should be loaded into
89 each well to allow visualisation but prevent smearing. If there are a large number of
different sized fragments (as occurs after restriction enzyme digestion) up to 20-30 pg
can be loaded without a significant loss of resolution.
• DNA ladders’ consisting of a mixture of DNA fragments of known sizes (Appendix
2) were loaded at either side of the sample wells to allow estimation of sample size at
the end of electrophoresis.
• The lid of the tank was closed. A voltage of 1-5 V/cm (measured as the distance
between the electrodes) was applied.
• When it appeared that sufficient separation of the DNA fragments had occurred (as
estimated by the progression of the coloured samples through the gel) the power pack
was disconnected and the gel removed from the electrophoresis tank. It was soaked
for 30-45 minutes in a solution of 0.5pg/ml EtBr in TAE and then placed over a UV
light source. The position of the PCR amplified nucleic DNA bands and the markers
could be seen where the EtBr intercalated between the bases. Each gel was
photographed to provide a permanent record of each experiment.
2.2 Results
A protocol for genotyping Min/+ mice has been described (Dietrich et al, 1993, Jacoby et al, 1996). Our protocol was based on this method but required modification as one of the primers, oIMR758, was different. It was 3 base pairs longer than its earlier counterpart, the extra bases being added to allow greater specificity. Conditions were altered on a step-wise basis in sequential experiments until the optimal conditions were obtained. This required a decrease in the final concentration of primer oIMR758, and increases in the annealing temperature, number of cycles and concentration of dNTPs. The final conditions are described (Appendix 1 and Section 2.1.2.1).
90 Ten untreated mice from two litters were genotyped initially, for use in setting up
experimental protocols of nucleic acid extraction PCR techniques and protein analysis. At
a later stage, 72 mice in 12 litters were genotyped. Of these 29 were Min/+ and 43 were
WT. Five male Mm/+ mice were kept aside for breeding purposes and 6 WT mice were
culled as they were excess to requirements. The rest were used in the trials of gene therapy. Figure 7 shows a typical triple primer genotyping PCR result on a 2% agarose
gel.
2.3 Discussion
Initial modifications to the Detrich protocol (Dietrich et al., 1993) took some time, but
once optimised the protocol described gave clear results. In the presence of any equivocal
results or the absence of good controls, the whole experiment was repeated.
91 Figure 6-Genotyping PCR Reactions
Site of Min mutation Apc cDNA
OIMR033 OIMR758 oIMR 034
Min/+ allele product,
300bp
WT allele product,
ÔOObp
Reverse primer oTMR034
5' 3'
2859 TTC CAC TTT CGC ATA AGO C 2841
Forward primer oIMR758
5' 3'
2529 TTC TGA GAA AGA CAG AAG TTA 2549
Forward primer oIMR033
5' 3'
2241 GCC ATC CCT TCA CGT TAG 2258
92 Figure 7-Example of Genotyping A Litter of Four Mice
1 2 3 4 5 6 7 9 10 11 12
500bp
Lane 1 1OObp ladder
Lanes 2 and 3 Positive Control samples from known Min/+ mice.
Both a 600bp PCR product and a 3OObp PCR product are amplified
Lanes 4 and 5 Positive Control samples from known WT mice.
A 600bp PCR product is amplified
Lane 6 Test sample 1 Min/+
Lane 7 Test sample 2 WT
Lane 8 Test sample 3 Min/+
Lane 9 Test sample 4 WT
Lane 10 Negative Control/ No DNA
Lane 11 Negative Control/ No reverse transcriptase
Lane 12 Negative Control/ No Supertaq
93 Chapter 3
Plasmid Preparation, Purification and Administration
94 3.1 Introduction
3.1.1 The Plasmid
The plasmid used in this project was APC pCMVp-Neo-Bam (Baker et al, 1990, Kinzler et al, 1991) (see Figure 8). This is based on one of the early plasmid vectors, pBR322, and contains both ampicillin and neomycin resistance cassettes. The APC gene has been cloned into the restriction enzyme site Bam HI and is under the control of a constitutively active CMV promoter. This plasmid has been widely used as it is able to undergo relaxed replication i.e. its replication is independent of cell division and cellular protein production. It does not replicate to high copy number (only 15-20 copies per cell) but expression can be increased by culture in chloramphenicol, which inhibits replication of the bacterial chromosome whilst allowing plasmid multiplication.
The pure plasmid (taken from a 50% glycerol stock stored at -80®C) was allowed to multiply in competent bacteria growing first on agar plates and then in culture medium.
When cell density in culture reached an optimal level the plasmid was purified using the
Qiagen Maxi Kit ( Qiagen Inc. Chatsworth, California, USA).
3.1.2 The Animals
All animal work was performed under the appropriate Home Office project and personal licences and with ethical committee approval from both University College London and the ICRF. Mice were treated at four weeks of age, prior to puberty, in order to precede any hormonal changes that might influence polyp progression. Each mouse was uniquely identified by litter, sex and ear markings. Following genotype analysis, at 21 days of age, the mice were kept in cages according to genotype, sex and litter. Following treatment, the cages were further split so that animals were only housed with others who had received the same treatment. 95 3.2 Materials and Methods
3.2.1 Transformation of Bacteria
The process of bacterial transformation by plasmid incorporation can be accelerated in some bacteria such as E.Coli. The cells are treated with an ice-cold salt solution (50mM calcium chloride) that alters the bacterial cell wall and increases DNA binding. Bacteria that have been treated in this way are termed competent. Rapid heating and recooling
(heat shock) then allows the DNA to be carried into the cell. For this project competent cells previously produced within the laboratory (XL 1-Blue) were used.
3.2.1.1 Protocol for Transformation of Competent Cells
• Three 200pl aliquots of competent cells were thawed on ice, (having been stored
frozen at -80°C) and transferred to 3 chilled micro frige tubes.
• Ipl of APC plasmid DNA (Ipg/pl) was added to the first micro frige tube and gently
mixed. (The number of transformants increases with the amount of DNA added until
the cells become saturated. In general 200 pi of competent cells are saturated by 4ng
ofDNA.)
• A control plasmid containing the p-galactosidase gene (pUC19) was added to the
second micro frige tube in the same way. This was the positive control.
• The third microfrige tube received no plasmid. This was the negative control.
• All cells were incubated on ice for 30 minutes.
• Each tube was heated at 45®C for 45-60 seconds and placed back on ice to cool for 2
minutes (this was the heat shock).
• 1ml of Luria-Bertani (LB) broth (Appendix 3) was added to each tube. The mixture
was incubated at 3TC for 45 minutes.
96 • The transformed cells were spun down in a 45° fixed angle bench top rotor centrifuge
at 14,000 rpm for 20 seconds, the supernatant was removed and the cells were
resuspended in 200pl of TrisEDTA (Appendix 3).
• 100-200pl of the transformation mix was plated onto LB agar plates (Appendix 3)
containing lOOpg/ml ampicillin and 50pg/ml X-gal. This was done in a laminar flow
hood using sterile equipment. A small flamed metal loop was used to spread the
transformed E. Coli onto the agar. The plates were incubated at 37°C overnight to
allow transformed cells (containing the ampicillin resistance cassette) to multiply.
• The next day colonies suitable for culturing were selected or the plates were wrapped
in Saran wrap and stored at 4°C.
3.2.2 Assessment of Transformation
Each bacterium normally takes up one piece of DNA during transformation. Degraded or linear DNA it is rapidly broken down and extruded by the host. Our aim was to identify those cells which had taken up whole plasmids containing a fimctional copy of the gene of interest.
E.Coli are sensitive to ampicillin, thus the negative controls (competent cells but no plasmid) did not grow on the agar / ampicillin plates. However, both plasmid vectors pUC19 and pBR322 contain an ampicillin resistance gene that allowed all the transformed cells to multiply overnight.
The positive control plasmid pUC19 contains the gene coding for the enzyme p- galactosidase. This enzyme metabolises the substrate 5-bromo-4-chloro-3-indolyl-p-D- galactoside (X-gal) to give a visibly blue product. X-gal was therefore incorporated into the agar of the positive control plates so that all transformed cells gave rise to blue 97 colonies. The number of blue colonies demonstrated the efficiency of the transformation. pBR322 carrying the APC gene contains both ampicillin and neomycin resistance
cassettes so bacteria which took up this plasmid were selected for in the presence of ampicillin.
3.2.3 Amplification of Transformed Cells
Bacteria being amplified purely as a source of plasmid DNA can be grown up in
'undefined’ medium, for example LB broth, which contains unspecified growth factors
such as tryptone or yeast extract.
3.2.3.1 Protocol
• Individual colonies from a freshly streaked selective agar plate were picked out (using
a sterile toothpick) and inoculated into 5ml of LB medium (with 1/1000 ampicillin) in
a 20ml universal container. This was incubated for 8 hours at 37°C with shaking. {E.
Coli cells divide every 20 minutes until the culture reaches a maximum density of 2-
3x 10^. This corresponds to a turbid appearance when viewed with the naked eye.)
The lid was loosely closed to allow oxygenation. Ampicillin was included to prevent
the growth of daughter cells that did not contain the plasmid of interest.
• This starter culture was then diluted 100 times in selective LB medium (containing
ampicillin) and cultured under identical conditions for a further 16 hours. This gave a
500ml cloudy suspension of approximately 10^ cells/ml, which was used for
preparation of larger quantities of plasmid DNA.
The growth rate and density of the culture can be monitored by measuring its optical
density (OD) at 600nm. For E. Coli, one OD unit corresponds to 0.8x10^ cells/ml at
600nm.
98 3.2.4 Purification of Plasmid
The Qiagen plasmid purification protocol is based on the principle of alkaline lysis of
bacteria (Birnboim and Doly, 1979) followed by binding of the plasmid DNA to an
anion-exchange resin under appropriate low-salt and pH conditions. RNA, proteins and
low molecular weight impurities are removed by a medium salt wash prior to elution of the plasmid in a high salt solution. Finally the plasmid is desalted and concentrated by
isopropanol precipitation.
3.2.4.1 Protocol
• Bacteria (approximately lOVml) were harvested from 250ml of culture medium by
centrifugation at 4,400 g for 15 minutes.
• Each pellet was resuspended in 10ml Buffer PI (Appendix 3). This contains a mildly
alkaline buffer. Tris (hydroxymethyl) methyl ammonium chloride, and RNAse A to
digest the RNA released during the procedure.
• 10ml of Buffer P2 was added (Appendix 3), the mixture was gently inverted 5 times
and left to incubate at room temperature for 5 minutes, until it appeared viscous. P2
contains sodium hydroxide (NaOH) and SDS. SDS solubilised the phospholipid and
protein components of the cell membrane and NaOH initiated alkaline dénaturation of
DNA and proteins. Alkaline lysis was restricted to 5 minutes in an attempt to release
plasmid DNA with minimal release of the larger chromosomal DNA.
• Chilled buffer P3 (Appendix 3) was added and mixed gently by inversion 5 times
(vortexing would shear chromosomal DNA). The mixture was then incubated on ice
for 20 minutes. P3 contains acidic potassium acetate which terminates the alkaline
lysis stage. The high salt concentration causes SDS to precipitate and the denatured
proteins, cellular debris and large chromosomal DNA become trapped in the salt-
99 detergent complexes. Plasmid DNA, being smaller and circular, renatures correctly
under acidic conditions and remains in solution.
• Precipitation of unwanted material was completed by centrifugation at ^20,000g for
30 minutes at 4®C. The supernatant was removed and centrifugation was repeated
under the same conditions in order to remove all cellular debris (which could block
the Qiagen-tip) and SDS (which could prevent DNA binding to the column).
• A Qiagen-500 tip (containing the anion-exchange resin) was equilibrated with 10ml
of Buffer QBT (Appendix 3). Detergent in the buffer (Triton X-100) reduces surface
tension in the tip so that the lysate will pass easily through it. It also established the
optimal salt and pH conditions for DNA to bind to the column.
• The pre-cleared lysate was applied and passed through the tip by gravity flow.
• The column was washed with 2x 30ml of Buffer QC (Appendix 3). This medium salt
buffer removed any remaining contaminants such as traces of RNA and protein.
• Plasmid DNA was eluted into a clean tube with 15ml of the high salt buffer QF
(Appendix 3).
• The DNA was precipitated with the addition of 0.7 volumes (11.5ml) of room
temperature isopropanol, gentle mixing and centrifugation at >15,000g for 30 minutes
at 4®C. (Solutions were at room temperature to avoid salt precipitation. Centrifugation
was carried out at 4®C to prevent overheating).
• The supernatant was removed, taking care not to displace the glassy pellet.
• The pellet was washed with 70% ethanol and centrifuged at >15,000g for 10 minutes.
(Ethanol removes precipitated salts and replaces isopropanol with a more volatile
agent that is easier to redissolve).
• The supernatant was discarded, the pellet air-dried for 10 minutes and re-dissolved in
200pl oflOxTE.
100 The plasmid was now ready for storage at -20”C or spectrophotometric quantitation to check its concentration and purity.
3.2.5 Spectrophotometric Quantitation of Nucleic Acids
The absorption of UV light by nucleic acid varies with the wavelength of the incident light and is maximal at 260nm. At this wavelength a solution with an optical density (OD) of one corresponds to a concentration of:
50pg/ml of ds DNA
40pg/ml of ss DNA or RNA
3.2.5.1 Protocol
• The spectrophotometer was switched on and allowed to warm up.
• The required wavelength of incident light was chosen.
• A 1ml sample of the buffer solution (TE) was placed in a cuvette in the light chamber.
The absorption of this standard was taken as zero.
• Each nucleic acid sample was diluted 1/1000 in TE and its absorption was measured.
• Calculations were performed using the formula shown.
[Nucleic Acid] pg/ml = OD 2 6 0 x dilution factor x concentration factor
The concentration factor is the concentration of the nucleic acid that gives an OD 2 6 0 of
one, as shown above.
To assess the purity of the sample the ratio of UV light absorption at 280nm incident
wavelength was also measured. Pure preparations have an OD 2 6 0 /OD2 8 0 value of 1.8 to
2.0. This ratio is reduced with contaminated samples. 101 3.2.6 Restriction Enzyme Digestion
Restriction enzymes bind to specific recognition sequences within dsDNA and break the nucleic acid strands at these points. The restriction enzymes Bam HI and Sal I have unique digest sites in the APC pCMVp-Neo-Bam plasmid, as shown in Figure 8. They were used to linearise the Maxiprep DNA product, to ensure that a plasmid of the appropriate size and sequence was produced and to exclude contamination by large amounts of non-specific cellular DNA. Their recognition sequences and optimum experimental conditions are shown in Appendix 3. All enzymes and buffers were purchased from Gibco BRL.
3.2.6.1 Protocol
• 1 pi of the plasmid Maxiprep product, was pipetted into a clean microfuge tube and
placed on ice.
• The appropriate restriction enzyme was added in excess. One unit of enzyme digests
1 pg of DNA to completion in 1 hour under optimum conditions. (5 units were used
here).
• 3 pi of the appropriate lOx restriction buffer was added to the mixture.
• The solution was made up to 30pl final volume with ddH20, and gently mixed by
pipetting up and down.
• The mixture was incubated in a waterbath at 37®C for 1 hour.
• After incubation, 5 and lOpl volumes of the digest product were run out on a 1%
agarose gel as described (Section 2.1.3.1.2), using a 1Kb DNA ladder.
102 3.2.7 Plasmid /Liposome Complex Formation
The plasmid was complexed with LIPOFECTAMINE™, a 3:1 weight: weight liposome formulation of the polycationic lipid 2,3-dioleoyloxy-N-[2-(sperminecarboxamido)ethyl]-
N,N-dimethy;-l-propanaminium trifluoroacetate (DOSPA) that complexes with DNA and the neutral helper lipid dioleoyl phosphatidylethanolamine (DOPE) in membrane filtered water (Life Technologies Inc, Gaithersburg, MD, USA).
3.2.7.1 Protocol
• APC pCMVp-Neo-Bam was mixed with LIPOFECTAMINE™, at room temperature,
in the ratio 1:12 by weight. The solution was made up to the final treatment volume
with sterile saline, and it was allowed to stand for a minimum of 15 minutes to allow
DNA/lipid complex formation.
• The control solution of 'LIPOFECTAMINE™ alone' consisted of an equal volume of
LIPOFECTAMINE^*^ to that used in the treatment arm, made up to the same final
volume with normal saline. It was allowed to stand for the same period of time prior
to use.
3.2.8 Rectal Treatment
Rectal treatment followed the treatment programme designed by Hargest et al (Hargest et
al., 1998). Overall, 21 mice from 6 litters were treated, at an age of 26 to 28 days.
3.2.8.1 Protocol
• 70pl o f plasmid DNA (at Ipg /pi) and 420pl LIPOFECTAMINE™ (at 2pg/ml) were
administered to each mouse treated with APC pCMVp-Neo-Bam, in a final volume of
500pl.
103 • The solution was drawn up in a 1ml syringe, a plastic mouse gavage tube was then
attached to the end of the syringe and all air bubbles were removed.
• An appropriate mouse was weighed and placed in a restrainer (normally utilised for
tail vein injections).
• The tail was elevated, K-Y jelly applied to the anus and the tip of the gavage tube was
inserted gently into the rectum.
• The entire volume of enema was slowly administered. Any leakage from the anus was
collected into a petri dish held beneath the mouse and re-inserted.
• Following treatment, the tail was held in the same position for approximately 1
minute, to keep the anus slightly elevated, and the mouse was then let loose.
• There did not appear to be any leakage after treatment so rectal plugs were not used
(to minimise discomfort).
• The animals were observed for twenty minutes and then replaced in their cages and
housed as normal until the day of death. Any ill animals were killed and an immediate
post-mortem was carried out.
• 9 Min/+ mice were treated with APC pCMVP-Neo-Bam.
• Three control groups were used:
• 4 WT mice were treated with APC pCMVp-Neo-Bam.
• 4 Min/+ mice were treated with 'LIPOFECTAMINE^’^ alone'.
• 4 WT mice were treated with 'LIPOFECTAMINE^’^ alone'.
3.2.9 Gavage Treatment
In line with the United Kingdom Co-ordinating Committee on Cancer Research
(UKCCCR) guidelines for animal welfare, the volume of fluid that can be administered to an animal by the oral route is limited to 2% of that animal's body weight. The mice used here were relatively small (13 to 19 grams). Therefore, the final treatment volume was 104 limited to 250|il to 300|il. The ratio of plasmid to LIPOFECTAMINE^'^ by weight (1:12) was maintained since this appeared to be the optimal ratio for efficient transfection in earlier studies (Hargest and Williamson, 1995). Thus the amount of plasmid delivered to each treated mouse by gavage was reduced to 35-42pg (rather than 70pg as used in the rectal study).
Overall, 20 mice from 4 litters were treated, at an age of 31 to 34 days.
3.2.9.1 Protocol
• The required treatment volume was drawn up in a 1ml syringe, a round tipped, metal,
mouse gavage tube was attached to the end of the syringe and all air bubbles were
removed.
• An appropriate mouse was weighed and then scruffed and held head upwards, so that
the body and head were immobilised.
• The gavage needle was inserted into one side of the mouth and advanced smoothly
until the stomach was entered. (Any distress was interpreted as mis-placement of the
syringe and the tube was withdrawn and replaced.)
• Once in position the treatment dose was smoothly administered, the tube was
withdrawn and the mouse was released.
• The animals were observed for twenty minutes and then replaced in their cages and
housed as normal until the day of death. Any ill animals were killed and an immediate
post-mortem was carried out.
• 8 Min/+ mice were treated with APC pCMVp-Neo-Bam.
• Three control groups were used:
• 4 WT mice were treated with APC pCMVp-Neo-Bam.
• 4 Min/+ mice were treated with 'LIPOFECTAMINE^'^ alone' 105 • 4 WT mice were treated with ’LIPOFECTAMTNE^’^ alone'.
3.2.10 Intra-peritoneal Treatment
Once again the treatment volume was limited to 2% of the animal's body weight. As the mice were small (15 to 17 grams), the final treatment volume was limited to 250pl in all cases. The ratio of plasmid to LIPOFECTAMINE™ by weight (1:12) was maintained. Thus the amount of plasmid delivered to each treated mouse by intra-peritoneal therapy was
35pg.
Overall, 12 WT mice from 4 litters were treated, all at an age of 31 days..
3.2.10.1 Protocol
• The required treatment volume was drawn up in a 1 ml syringe, a 25 gauge needle was
attached to the end of the syringe and all air bubbles were removed.
• An appropriate mouse was weighed and then scruffed and held head upwards, so that
the body and right hind leg were immobilised.
• The needle was inserted just to the right of the midline, approximately half way down
the abdomen, and advanced smoothly until the peritoneum was punctured.
• The treatment was inserted slowly and the needle withdrawn. The mouse was then
released.
• The animals were observed for twenty minutes and then replaced in their cages and
housed as normal until the day of death. Any ill animals were killed and an immediate
post-mortem was carried out.
• 8 WT mice were treated with APC pCMVp-Neo-Bam.
• 4 WT mice were treated with 'LIPOFECTAMINE^^ alone', as a control group.
106 3.3 Results
3.3.1 Plasmid Preparation
Initial attempts at amplification of APC pCMVp-Neo-Bam from previously transformed cells gave poor results. However, transformation of new competent cells with plasmid stock gave a better starting point. Once a good 500ml culture of transformed bacteria was ready, the Maxiprep purification protocol was straight forward to use, although the final plasmid yield (as measured by spectrophotometry) was variable, ranging from 1.04mg/ml to 3.25mg/ml. All samples were diluted to a constant concentration of 1 mg/ml for end use. The purity of the preparations was consistently high.
3.3.2 Restriction Enzyme Digestion
The enzymes Bam HI and Sal I were used to digest plasmid DNA under the conditions described (Appendix 3). The APC gene insert, approximately 8.9Kb in size, has been cloned into the Bam HI site. Therefore, cleavage with Bam HI should give rise to a linearised vector of approximately 6.6Kb and a free 8.9Kb insert. Sal I cuts the vector away from the insertion site and should produce a single 15.5Kb fragment.
Figure 9 shows the results of the restriction digest run out on a 1% agarose gel, stained with EtBr and photographed. The expected fragments were produced with no other
contaminating DNA.
3.3.3 Plasmid Administration
The results of the treatment are covered in detail in Sections 4.2, 5.2 and 6.3.2.
107 3.4 Discussion
3.4.1 Plasmid Preparation
Insertion of human cDNA into a newer plasmid vector would have allowed a higher
copy number during plasmid amplification in bacteria, and a greater overall yield of plasmid. However, the manufacture of a new vector was beyond the scope of this project.
Instead, the process of plasmid preparation and purification was repeated many times, to
give enough to treat the required number of animals.
3.4.2 Restriction Enzyme Digestion
The results of the restriction digest, combined with the results of spectrophotometry,
confirmed that the Maxiprep was producing a pure preparation of the plasmid of interest.
108 Figure 8-The APC pCMVp-Neo-Bam Plasmid
APC cD N A
am HI Xba I Sal I
750 bp 4 ► 560bp
Intron
Neomycin Resistance Eco RI
Hind III N^Ampicillin Resistance ^
Sca I I Restriction Digest Sites (Bam Hl, Sal I, Xba I, Hind III, Eco RI and Sca I) I Origin of Replication CMV promoter
Rabbit (3-gIobin
Poly A Adapted From Baker et al 1990
109 Figure 9-Restriction Enzyme Digestion of Purified
APC pCMVp-Neo-Bam
W 12 Kb 9 Kb
Lanes 1 and 5 1 Kb ladder
Lane 2 Undigested APC pCMVp-Neo-Bam. Cireular DNA
migrates slowly and remains near the top of the gel
Lane 3 Bam HI digest of APC pCMVp-Neo-Bam gives an 8.9Kb
APC cDNA insert and a 6.6 Kb plasmid skeleton
Lane 4 Sal I digest of APC pCMVp-Neo-Bam gives a linear
15.5Kb DNA
10 Chapter 4
Study of Transgene Expression Following Gene Therapy
111 4.1 Methods
This part of the project attempted to document the physical extent of transgene expression
after a single dose was delivered to the rectum or stomach of Min/+ mice (and WT
controls) and to follow this over a four day time course. Having considered the possibility
of intra-peritoneal APC gene therapy in the treatment or prophylaxis of desmoids, the
extent of transgene expression following a single intra-peritoneal injection of APC
pCMVp-Neo-Bam was also recorded, using only WT mice (produced in excess in the
Min/4- breeding programme).
Transgene expression is best studied at the mRNA level as this identifies those copies of
the transgene that have not only entered the cell but have initiated transcription. This is
achieved by extraction of total mRNA fi*om a tissue, followed by specific amplification of
the message of interest (by reverse transcriptase PGR)
4.1.1 Organ Retrieval
The mice were killed on days 1, 2, 3 and 4, the date of treatment being termed Day 0.
Each day two ’treated' mice were killed along with 1 mouse fi*om each of the control
groups described (see Tables land 2). Each mouse was placed in an airtight glass carbon
dioxide chamber and the gas flow was increased until the mouse lost consciousness and
died. A prolonged period (one minute) of stillness and absence of reflex response to pain
was taken to indicate death.
Each mouse had an identical post-mortem. When pinned out on a board, the abdomen was
carefully opened and inspected to exclude the presence of fi’ee intra-peritoneal fluid,
intestinal distension or perforation. The liver, spleen, kidneys and pancreas were also
inspected and any abnormality was recorded.
112 The entire gastro-intestinal tract was removed. It was divided into stomach, small bowel sections one, two and three (numbered from proximal to distal small bowel in approximately equal sections), colon (as far as the splenic flexure) and recto-sigmoid
(everything distal to the splenic flexure). Each segment was opened longitudinally, washed in cold PBS (to remove faecal matter), and examined in good light under a xlO dissecting microscope to note the position and size of any macroscopically visible polyps.
Samples were then snap frozen in liquid nitrogen in appropriately labelled cryogenic tubes. Tubes were labelled according to the date, the experiment, the genotype and individual identification number of the mouse, and the body part contained. Polyps were dissected out under the microscope and frozen separately. The liver, kidneys and gonads were also removed and snap frozen. Following intra-peritoneal treatment, sections of peritoneum were recovered.
Following initial freezing in liquid nitrogen, all tissue samples were transferred to a -80°C freezer until required.
4.1.2 Preparation of mRNA
A commercially available kit, QuickPrep® Micro mRNA purification kit (Pharmacia
Biotech Inc.) was used. This allowed the rapid production of polyadenylated RNA from multiple small samples (typically 20pg) without the need for purification of total cellular
RNA. The kit produced virtually pure RNA, with minimal DNA or protein contamination. However, the mRNA yield was not high, typically being under 3pg per sample. This limited the number of cDNA copies produced from each tissue sample.
All reactions were carried out at room temperature unless otherwise stated. In order to prevent sample contamination with exogenous ribonucleases, gloves were worn at all
113 times. Bench surfaces and Gilson pipettes were cleaned prior to use and all plasticware coming into contact with the samples was double autoclaved (Sambrook et al., 1989a).
Diethyl pyrocarbonate (DEPC) treated water (which is RNase free) was used to clean the tissue homogeniser between samples.
Tissue samples were removed from storage at -80°C and thawed on ice. Nucleic acid was extracted by rapid homogenisation on ice in a buffered solution of guanidinium thiocyanate (GTC) that inactivates endogenous tissue RNases (Chirgwin et al., 1979). The extraction buffer was then diluted in elution buffer to give a final concentration of GTC low enough to permit hydrogen bond formation between the poly-A tract of mRNA and the oligo(dT)-cellulose used to purify it, but high enough to maintain inhibition of RNase activity. Cellular proteins began to precipitate at this stage, and were removed by centrifugation. The cleared tissue homogenate was mixed with oligo(dT)-cellulose and purified by sequential washes in high and then low salt buffers. The initial high salt washes removed the majority of contaminating proteins and nucleic acids, later washes completed the removal of carbohydrates. Finally, the mRNA was eluted from the column with pre-warmed elution buffer.
4.1.2.1 Protocol
4.1.2.1.1 Preparation
• The kit was stored at 4°C but removed from the fridge approximately 30 minutes prior
to use, to equilibrate to room temperature.
• The extraction buffer (Appendix 4) was placed in a waterbath at 37®C for ten minutes,
to dissolve any crystalline material. Thereafter it was cooled to room temperature.
• 0.5ml of elution buffer (Appendix 4) per sample was warmed to 65°C in a waterbath,
for use in the final step of elution of mRNA. 114 The oligo(dT)-cellulose solution (Appendix 4) was mixed well to re-suspend the
particles in buffer and 1ml aliquots were placed in 1.5ml screw top microfuge tubes,
labelled according to the tissue samples being studied.
4.1.2.1.2 Nucleic Acid Extraction
• Tissue samples (typically 15-20pg) were removed from the -80®C freezer, chopped
into three or four pieces and placed immediately in a 10ml round bottomed tube with
400pl extraction buffer.
• The tube was placed in a container of ice and the tissue was rapidly homogenised,
until a uniform tissue suspension was produced.
• 800pl elution buffer was added to the tube and the sample was briefly homogenised
once again.
• The extracted tissue sample was transferred to a 1.5ml screw top eppendorf tube and
centrifuged at 13,000g for one minute. At the same time a microfuge tube containing
oligo(dT)-cellulose was centrifuged.
4.1.2.1.3 Binding
• The buffer from the oligo(dT)-cellulose was removed and 1ml of the clarified tissue
extract was placed on top of the cellulose pellet. The tube was closed and inverted
manually to re-suspend the oligo(dT). The tube was then placed in a rotating wheel
for three minutes to facilitate hydrogen bond formation between the poly-A of the
mRNA and the oligo(dT) tract on the cellulose.
• The sample was centrifuged at 13,000g for ten seconds.
• The supernatant was aspirated and discarded.
115 4.1.2.1.4 Washing
• The oligo(dT) pellet was washed 5 times with 1ml of high salt buffer (Appendix 4),
and then twice with 1ml of low salt buffer (Appendix 4). The pellet was centrifuged at
13,000g for five seconds between each wash.
• The oligo(dT)-cellulose was re-suspended in 0.3ml low salt buffer and transferred to a
polypropylene Micro spin™ column (Pharmacia Biotech Inc.), in a sterile microfiige
tube.
• The microfuge tube and column were centrifuged at 13,000g for five seconds and the
effluent was discarded.
• The concentrated oligo(dT) pellet was then washed twice with 0.5ml of low salt
buffer.
4.1.2.1.5 Elution Step
• The column was placed in a new sterile microfuge tube and 0.2ml of the pre-warmed
elution buffer (65°C) was added to the top of the resin bed.
• The tube and column were centrifuged at 13,000g for five seconds. The eluate
produced was the purified mRNA.
• In order to maximise the yield the elution step was repeated. This gave a slightly
greater overall quantity of mRNA but a lower concentration.
4.1.2.1.6 Concentration of mRNA
• The tube containing the purified mRNA was placed on ice.
• lOpl glycogen solution (Appendix 4), 40pi potassium acetate solution (Appendix 4),
and 1ml of 95% ethanol (chilled to -20°C) were added to the 400pl eluate.
• The sample was placed at -20®C for a minimum of 30 minutes, preferably overnight. 116 • The precipitated mRNA was collected by centrifugation for five minutes at 4°C and
13,000g.
• The supernatant was removed by inverting the tube over a clean paper towel on the
bench-top and tapping the open edge of the tube gently on this surface.
• The precipitated mRNA was re-dissolved in an appropriate volume of elution buffer,
normally 40pi, one tenth the original volume.
• This was stored at -80°C or placed on ice for immediate use in the production of a
cDNA copy.
4.1.3 Preparation of cDNA
Messenger RNA samples were converted into cDNA using the commercially available
First-Strand cDNA Synthesis Kit (Pharmacia Biotech Inc.). Here, Moloney Murine
Leukaemia Virus reverse transcriptase (RT) was used to catalyse the synthesis of first strand cDNA, in the presence of a random hexanucleotide primer, pd(N)6. The
RNA/cDNA heteroduplex was then used as the template for DNA polymerase in PCR reactions (Saiki et al., 1985, Mullis and Faloona, 1987).
All precautions were taken to minimise RNA degradation and sample contamination, as described above. In addition, all reactions were carried out on ice unless otherwise stated.
4.1.3.1 Protocol
• The kit was stored frozen at -20°C. Components were removed from the freezer and
placed on ice prior to use.
• The RT (Appendix 5) was gently pipetted up and down to resuspend particulate
matter.
117 • 5|al of RT, l|il of pd(N)6 and Ipl of dithiothreitol (DTT) were placed in each
microfuge tube.
• Concentrated mRNA (in 40pl elution buffer) from each tissue sample was denatured
by heating to 65°C for ten minutes. It was then replaced on ice.
• 8pl of this mRNA was placed in a microfuge tube containing the reaction mix, and
the tube was labelled according to the RNA sample added.
• Three identical cDNA reactions were set up for each mRNA sample. The remaining
mRNA was stored at -80°C.
• Negative control tubes, with no mRNA or no RT, were also prepared for each
experiment. Their final volume was made upto 15p,l with DEPC water.
• Each tube was mixed gently prior to incubation in a waterbath at 37°C for one hour, to
allow reverse transcription to occur under optimal conditions.
• The microtubes were removed from the waterbath and placed at 90®C for five minutes
to inactivate the RT.
• The resulting RNA/cDNA complex was used for amplification in the PCR reaction or
stored at -20°C.
4.1.4 Amplification of cDNA by The Polymerase Chain Reaction
The principles of PCR were explained in Section 2.1.2.
The primers utilised here were:
PI 5’ GGATCCTGAGAACTTCAG 3' (552bp-569bp)
P2 5' ATG CTT GTT CTG AGA TGA C 3’ (750bp-768bp)
Primer PI is complementary to the rabbit P-globin sequence of APC pCMVp-Neo-Bam
(see Figure 8), just upstream of a splice donor site. Primer P2 spans part of the human
APC sequence. Used together they amplify a 1396 base pair fragment with unspliced 118 amplification and a 796bp message following in vivo splicing (Westbrook et al, 1994).
Thus, where the plasmid was taken into cells and transcribed, a band at either or both these molecular weights was expected following cDNA amplification in the presence of these primers.
To provide a positive control, purified plasmid DNA (at a concentration of approximately
0.2pg) was amplified with the same primers and the same cycling conditions for each
experiment.
To ensure that absence of the appropriate PCR signal was due to lack of transgene
expression and not simply failure to produce high quality mRNA or cDNA, or to add the
correct reaction constituents, a simultaneous control PCR reaction was run. The third
cDNA copy of each mRNA sample was amplified with primers for the ubiquitously
expressed mouse p-actin. Use of the primers MBAF and MBAR produced an 872 bp
fragment.
MBAF 5' CCC ATC TAC GAG GGC TAT GC 3' (570bp-589bp)
MBAR 5' TTT GCT CCA ACC AAC TGC TG 3' (1442bp-1423bp)
Transgene expression was confirmed when two cDNA copies from an individual tissue
sample gave a 1396bp and/or a 796bp fragment following amplification with primers PI
and P2, in the presence of good negative controls (no signal in the no cDNA and no RT
samples).
Transgene expression was excluded when no appropriately sized PCR product was
generated from two cDNA copies of a tissue sample, in the presence of good positive
119 controls (appropriate size fragments generated from plasmid amplification with PI and P2 and from matched cDNA amplified with MBAF and MBAR).
4.1.4.1 Protocol
• Two cDNA samples (15pl each) from each tissue sample were was spun down and
placed in PCR reaction tubes with:
5pi 1 OpM stock PI (1 pM final concentration)
5 pi 10pM stock P2 (1 pM final concentration)
2.5 units Supertaq polymerase
• The third cDNA sample was spun down and placed in a PCR tube with:
1.5pl 1 OpM stock MBAF (0.3pM final concentration)
1.5pi 1 OpM stock MBAR (0.3pM final concentration)
2.5 units Supertaq polymerase
All tubes were made up to 50pl final volume with ddHaO. dNTPs and buffer did not need to be added as they were already contained within the cDNA mixture.
• The reaction mixture was covered with mineral oil to prevent evaporation during the
dénaturation phase.
• Positive control tubes containing approximately 0.2pg purified plasmid were included
in all experiments (Appendix 5).
• Negative control tubes lacking cDNA or RT (produced when the cDNA was made)
were carried through in all PCRs.
• The PCR programme was run as shovm (Appendix 5).
• 12.5pi of each PCR product was mixed with 2pl Orange G loading dye and loaded
onto a 2% agarose gel for analysis.
• A lOObp DNA ladder (5 pi) was run on each gel to estimate the size of amplified
bands. 120 4.2 Results
4.2.1 Transgene Expression Following Rectal Treatment
Twelve mice (8 Min/+ and four WT) were initially treated with APC pCMVp-Neo-Bam.
One of the Min/+ mice became non-specifically unwell on day one and was humanely sacrificed. (Another member of the same litter who had not been involved in the experiment developed the same symptoms and was also killed.) Post-mortem failed to reveal any abnormality related to treatment (see Section 5.2.1), but the tissue was not processed fast enough to be used in the study, so an additional Min/+ mouse from the same litter was treated with plasmid and taken out to four days to keep up the study numbers.
4.2.1.1 Recto-sigmoid Samples
Overall, ten out of the twelve healthy mice treated with APC pCMVP-Neo-Bam displayed exogenous plasmid mRNA in the 'recto-sigmoid' samples (see Figures 10 and
13). Of the Min/+ mice treated with plasmid, only one of those killed after two days and one of those killed after three days failed to express the transgene. All of the WT mice treated with plasmid expressed the transgene in their 'recto-sigmoid' tissue up to four days after treatment. Both spliced and unspliced PCR products (1396bp and 796bp respectively) were seen. An additional band between these sizes was sometimes present.
This was not amplified from the plasmid control. None of the eight mice treated with
'LIPOFECTAMINE™ alone' showed a positive band after rtPCR from the 'recto-sigmoid' samples.
Rectal samples from all the mice (treated with plasmid or 'LIPOFECTAMINE^’^ alone') expressed p-actin mRNA.
121 Ethical committee approval restricted the number of mice used in each trial. However,
WT mice bred in excess, in order to obtain an adequate number of Min/+ animals, were available for use. Following the observation that transgene expression was maintained in rectal tissue four days after treatment, two ftirther WT mice were treated with the plasmid enema and taken out to day 5 and 6. Neither rectal sample tested positive for transgene mRNA.
4.2.1.2 Colonic samples
'Colonic' samples (ascending and transverse colon up to the region of the splenic flexure) fi*om all the mice were analysed for the presence of plasmid mRNA. There were no positive results. Figure 11 shows the it PCR results fi*om the colonic samples of those mice who had expressed the transgene in their rectal tissues.
4.2.1.3 Liver and Gonads
Liver and gonadal samples of all mice were also analysed. No 1396bp or 796bp PCR ft-agments were produced. However, low molecular weight fi*agments were seen following rtPCR fi-om the liver samples of some of the animals treated with plasmid, (see
Figure 12). These were not present in the mice treated with 'LIPOFECTAMINE^^ alone'.
4.2.1.4 Controls
In all experiments, control animals (treated with 'LIPOFECTAMINE™ alone') tested negative for the exogenous mRNA. rtPCR results were only accepted if the appropriate positive and negative controls were present (described in Section 4.1.3 ).
122 4.2.2 Transgene Expression Following Gavage
4.2.2.1 Stomach and Small Bowel Samples
Only four out of twelve mice treated with APC pCMVp-Neo-Bam, expressed the plasmid mRNA in gastric tissue. Three of these mice (all Min/+) also expressed the exogenous message in the most proximal part of the small bowel. Only unspliced PCR products were seen (Figures 14, 15 and 18). There was no evidence of transgene expression in more distal segments of small bowel.
Stomach and small bowel samples from all the mice expressed p-actin mRNA.
4.2.2.2 Liver and Gonads
No plasmid specific 1396bp or 796bp PCR fragments were produced from either the liver or the gonads. However, faint bands of approximately 300bp and 400bp were produced from liver and gonadal samples of some of the mice, including those treated with
'LIPOFECTAMINE™ alone' (see Figure 16, 17 and 18).
4.2.2.3 Controls
No samples taken from mice treated with 'LIPOFECTAMINE^^ alone' tested positive for transgene mRNA. rtPCR results were only accepted in the presence of appropriate positive and negative controls.
4.2.3 Transgene Expression Following Intra-peritoneal Treatment
4.2.3.1 Peritoneal and Mesenteric Samples
Of the eight WT mice treated with APC pCMVP-Neo-Bam, all animals expressed the transgene in the peritoneum, and only one mouse failed to express the exogenous mRNA in its small bowel mesentery. Both spliced and unspliced PCR products were seen in the 123 mesentery, but only full length mRNA was present in the peritoneal tissue. None of the four mice treated with 'LIPOFECTAMINE^’^ alone' showed a positive band (see Figures 19 and 20 and Table 3).
Peritoneal and mesenteric samples from all the mice (treated with plasmid or
'LIPOFECTAMINE™ alone') expressed P-actin mRNA.
4.2 3.2 Intestinal Tissues
Small bowel, colon and rectal samples from animals treated with APC pCMVp-Neo-Bam showed a good level of transgene expression, with no obvious decrease in expression from the first to the fourth day after treatment (see Figure 20 and Table 3).
4.2.3 3 Liver and Gonads
Transgene expression was seen in the liver of seven of the eight mice treated with the plasmid (see Figure 20 and Table 3). Additional low molecular weight bands were one again present after rtPCR from the livers of animals treated with APC pCMVp-Neo-Bam.
They were not seen in mice who received 'LIPOFECTAMINE™ alone'.
Over a third of the mice treated with APC pCMVP-Neo-Bam also showed transgene expression in the gonads (see Figure 20 and Table 3). All three of these mice were male.
4.2.3 4 Controls
Once again, rtPCR results were only accepted in the presence of appropriate controls and animals treated with 'LIPOFECTAMINE^’^ alone' consistently tested negative for exogenous mRNA.
124 4.3 Discussion
4.3.1 Preparation of mRNA
Initial experiments with the QuickPrep® Micro mRNA purification kit demonstrated that it gave a pure RNA product (as assessed by spectrophotometry) but the relatively low yield necessitated concentration of the end-product in a reduced volume of elution buffer prior to use.
4.3.2 Preparation of cDNA
Judging by the control PCR amplification of p-actin, the First Strand cDNA Synthesis Kit also gave consistently good results. The occasional failure of an entire experiment was put down to failure to add the RT or a sub-optimal incubation temperature.
4.3.3 Amplification of cDNA by the Polymerase Chain Reaction
4.3.3.1 Primers For Amplification of AFC pCMVp-Neo-Bam
Two sets of primers have been described for the detection of expression of an APC transcript after transfection with APC pCMVp-Neo-Bam (Hargest et al., 1998,
Westbrook et al., 1994). Both were tested in preliminary experiments. However, the former set did not give reliable results, with frequent false positive signals in samples taken from untreated Min/+ mice. This may have been due to the high homology between murine and human APC or as a result of previous use of these primers in the laboratory
and non-specific contamination of stock solutions with amplified PCR products.
Therefore, the primers described by Westbrook et al (referred to here as PI and P2) were used. These gave no false positive results in untreated Min/+ or WT mice when the negative controls were clear.
125 4.3.4 Transgene Expression Following Rectal Treatments
Rectal administration of APC gene therapy is a well-established technique in rodents
(Westbrook et al, 1994, Hargest et al., 1998, Arenas et al., 1996). In human disease, suppositories and enemas are commonly used in the management of inflammatory bowel disease or troublesome constipation and are popular on the continent as an alternative route for the administration of analgesic agents. Enema formulations have been produced that allow self-administration of drugs with minimum discomfort.
Individuals with FAP, the potential target group for APC rectal gene therapy, have a strong incentive to use topical applications that might reduce their risk of rectal carcinoma in the rectal stump, or postpone the need for further surgery after an initial
IRA.
4.3.4.1 Recto-sigmoid Samples
More than 80% of mice treated with APC pCMVp-Neo-Bam expressed the transgene in the recto-sigmoid. Individual animals not expressing the exogenous APC mRNA may have evacuated their bowels soon after receiving the enema since no attempt was made to plug the rectum. Therefore, absence of gene expression may have been due to failure of administration rather than failure of transfection. This should not be an issue in human gene therapy as treatment could be administered after spontaneous defecation or re administered in the event of unforeseen loss of the enema.
The transgene was expressed in rectal tissue for a period of at least four days. In an attempt to look further into the duration of transgene expression, two additional WT mice were treated with APC pCMVp-Neo-Bam and taken out to 5 or 6 days. A positive test would have suggested prolonged expression of the transgene, but failure of expression in
126 an individual animal at each time point was not able to exclude a gradual decrease in expression with time. If the optimal treatment interval in human disease is one that maintains a consistently high level of transgene expression, four days is probably suitable.
The PCR product occasionally seen between 1396bp and 796bp may have been a splice variant or due to non-specific amplification of genomic DNA fragments similar but not identical to the target sequence.
4.3.4 2 Colonic Samples
It appears that transgene uptake is limited to the local area of the recto-sigmoid and the gene is not carried more proximally in the bowel. This is probably not a drawback to human gene therapy as rectal treatment has only been proposed as treatment for the rectal stump following IRA. This is a short segment of bowel, easily reached by enema preparations.
4.3.4.S Liver and Gonads
Complete absence of transgene expression in the liver or gonads suggests that rectal delivery of the transgene does not favour entry into the systemic circulation in significant amounts. The absence of transgene expression in the gonads following local treatment is a useful safety feature in a prophylactic treatment aimed at young individuals who may wish to procreate.
Liver samples in all three trials demonstrated unexpected PCR products at low molecular weights. These were only seen in samples from mice treated with APC pCMVp-Neo-
Bam in the rectal and intra-peritoneal trials. However, they were also present in the control ('LIPOFECTAMINE^^ alone’) samples of mice treated by gavage. Therefore, it is
127 likely that all of these bands are due to non-specific amplification of genomic DNA that has a high degree of homology with one or both primers.
4.3.5 Transgene Expression Following Gavage
4.3.5.1 Stomach and Small Bowel Samples
Transgene expression in the stomach and proximal small bowel of the mice gavaged with
APC pCMVp-Neo-Bam was low (4/12 and 3/12 respectively) when compared to expression in the recto-sigmoid samples of 10/12 animals treated rectally. The decreased dose of plasmid administered by gavage may have had a role to play. The decrease in treatment volume was probably not as important given that the stomach mucosa is not as tightly convoluted as the rectum so most gastric mucosa would have been exposed to the therapy.
The transfection efficiency of cationic lipids is dependent upon the cell type transfected
(Caplen et al., 1995b). It is possible that upper gastro-intestinal cells were simply not as responsive to transfection with this plasmid as the colonic epithelium. The physical surroundings of the plasmid in the upper gastro-intestinal tract may also have reduced transfection efficiency. The thick mucus layer in the stomach (protecting it from acid digestion) could have reduced direct exposure of the gastric mucosa to the plasmid. Once in the small intestine, lipases in pancreatic and intestinal secretions may digest a proportion of the lipid molecules leaving the associated DNA unprotected in the presence of pancreatic deoxyribonucleases.
In vitro studies suggest that transfection of cell lines is less effective at extremes of pH
(Caplen et al., 1995b). Although this may be due to a direct effect of pH on cell lines in culture, an adverse effect on the plasmid/ lipid complex cannot be ruled out. Exposure of
128 the APC pCMVp-Neo-Bam plasmid/ LIPOFECTAMINE™ complex to strong acid (pH<2) in the stomach may have initiated complex degradation prior to transfection. Alteration in the charge of the lipid moiety and subsequent change in the plasmid/lipid interaction may also have occurred.
Bile salts act in a physiological manner to cause émulsification of lipids and to enhance lipid uptake into small intestinal cells through the formation of micelles (bile salt lipid complexes which are highly soluble due to the charge of the bile salts). The overall effect of bile salts on the DNA/lipid complexes used in this project is unclear. It could lie anywhere between complex destruction and enhanced intestinal uptake of the vector and plasmid. Work in this group is on-going looking at the effect of bile in vitro on upper intestinal cell transfection efficiency.
The absence of transgene expression more distally in the small bowel mirrors the local effect seen in rectal gene therapy. Once again this should not be a significant concern in human disease where the majority of troublesome polyps are proximal within the duodenum, clustered around the ampulla of Vater (Spigelman et al., 1989).
4.3.S.2 Liver and Gonads
No transgene expression was seen in the liver or gonads following gavage with APC pCMVp-Neo-Bam. This supports the suggestion that gastro-intestinal gene therapy does not lead to systemic circulation of the transgene.
Non-specific rt PCR products seen in the liver and gonads were put down to non-specific amplification of genomic DNA homologous to the primer sequences.
129 4.3.6 Transgene Expression Following Intra-peritoneal Treatment
Intraperitoneal chemotherapy has been widely used alone and in combination with intravenous agents in the adjuvant treatment of ovarian cancer (Morgan et al, 1999).
There have also been occasional reports of its use in the management of peritoneal métastasés from other origins, such as appendiceal or colorectal tumours (Horsell et al,
1999). Treatment is generally administered at intervals over a prolonged period, through an intraperitoneal catheter. The time taken for the drug to cross the peritoneum into the circulation means that high doses of chemotherapeutic agents can be administered locally into the peritoneal cavity and moderate doses are delivered slowly to the systemic circulation without a toxic peak (Gladieff et al, 1999). However, there are technical problems, including intraperitoneal sepsis and catheter occlusion (Horsell et al, 1999).
Intra-peritoneal gene therapy is a step on from intraperitoneal chemotherapy. Delivery of liposome///*SFitA: complexes has been used in animal models of pancreatic carcinomatosis.
The transgene was taken up and expressed in 10% of peritoneal tumour cells (Aoki et al,
1997). Tumour cells preferentially expressed the transgene, possibly due to an increased proliferation rate (Kikuchi et al, 1999). Intra-peritoneal administration of corrective gene therapy using WT p53/liposome complexes has been used in a mouse model of residual glioblastoma. Good levels of tumour transfection and cell death were seen (Hsiao et al,
1997). However, neither of these studies recorded transgene expression in 'normal' tissues within the abdominal cavity.
Intra-peritoneal gene therapy should theoretically give rise to local effects and/or systemic distribution of the complexes following absorption into the circulation. It is known that intra-peritoneal liposomes are taken up and destroyed by peritoneal macrophages (Huong et al, 1998) and lipid/DNA complexes within the systemic
130 circulation are rapidly degraded, especially in the presence of the neutral lipid DOPE that is found in LIFOFECTAMTNE™ (Li et al., 1999). Thus part of this project considered the transfection efficiency in various tissues following intra-peritoneal administration of APC pCMVp-Neo-Bam in WT mice.
In the absence of any effective prophylaxis or treatment for intra-abdominal desmoid disease, corrective gene therapy with WT APC may be a way to 'stabilise’ the at risk connective tissue (such as the small bowel mesentery and peritoneum) that appears prone to tumour formation following 'stressful' interventions such as surgery. It could be administered on a regular basis in high-risk families or around the time of surgery in others with 'classical FAP'.
4.3.6.1 Peritoneal and Mesenteric Samples
The majority of desmoids in FAP are intra-abdominal (70%). Of these 40-50% arise in the mesentery (Einstein et al., 1991, Soravia et al, 2000) but some may arise from peritoneal fibroblasts. High levels of transgene expression (7/8 and 8/8 animals respectively ) were seen in both of these tissues and the level of expression was consistent out to four days after treatment. The next step in evaluation of intra-peritoneal APC gene therapy as a means of treatment for desmoids would be to extend the period between treatment and tissue collection and then to look at the effect of treatment in an animal model of desmoid disease. The Min/+ mouse is not a good model for desmoid disease but others have been described. The most commonly used is the ^/?cl638N mouse (described in Section 1.8.2). However, this has largely musculo-skeletal desmoids rather than intra abdominal disease.
131 Studies described above (Aoki et al, 1997, Kikuchi et al, 1999, Hsiao et al., 1997)
suggested that intra-peritoneal gene therapy favours transgene expression in peritoneal
tumour deposits rather than in 'normal* tissue. This study clearly demonstrates transgene
expression in benign mesentery and peritoneum. Considering the prevention of desmoid
disease, expression of WT APC in pre-neoplastic lesions might prevent tumour initiation.
However, even if neoplastic change occurred the increased rate of proliferation seen in
locally aggressive disease might itself enhance transgene expression in affected cells
(Tseng et al., 1999).
4.3.6.2 Intestinal Tissues
Intra-peritoneal administration of gene therapy is a potential means of delivering
intestinal gene therapy. Five to six out of eight mice treated with the plasmid expressed
the transgene in the small bowel, colon and rectum. However, rtPCR was performed on
sections of whole bowel in the treated animals and it was not clear whether transgene
expression was confined to the serosa or reached to the mucosa where its effect might be
beneficial. Mucosal stripping and rtPCR of just this segment of bowel would be useful in
future experiments to consider this problem.
Prolonged or repeated intra-peritoneal administration of gene therapy in Min/+ mice
could also be attempted to record the clinical effects of such treatment on intestinal
disease.
4.3.6.3 Liver and Gonads
The high level of transgene expression in the liver seen here (7/8 treated animals) was
probably due to a local effect of transfection into superficial hepatocytes. High level gene
expression in liver following systemic administration of plasmid/ lipid complexes has
132 been attributed to trapping of blood within sinusoids rather than transfection of parenchymal cells (Griesenbach et al, 1998). However, systemic administration of plasmid/ lipid complexes has been shown to cause low level transgene expression in the liver even after exsanguination of mice prior to tissue collection (Templeton et al., 1997).
The additional low molecular weight PCR products seen in some liver samples fi*om mice treated with the plasmid were once again thought to be due to non-specific amplification of genomic DNA.
Interestingly, transgene expression was seen in gonadal samples of three male mice treated with intra-peritoneal APC pCMVp-Neo-Bam. It is possible that the treatment initiated a rise in intra-abdominal pressure, allowing fluid to move directly through a patent processus vaginalis and come into contact with the testes. Alternatively, gonadal expression could have been secondary to systemic circulation of the complex. The over representation of transgene expression in male mice might lead one to prefer the former explanation, however only two of the eight mice treated with the plasmid were female so
the likelihood of seeing gonadal expression in a female was low.
If intra-peritoneal gene therapy were to be used in the management of human disease
ftorther studies of gonadal expression would be needed so that adequate advice could be
given to patients.
Overall the level of transgene expression in all tissues following intra-peritoneal gene
therapy was higher than might have been expected. This is a method ofAPC gene therapy
that requires more extensive investigation, particularly in relation to its use in the
prevention and management of desmoid disease.
133 Tables 1 and 2-Protocols For Animal Euthanasia
Table 1 Rectal And Gavage Trials
Treated with Treated with
APC pCMVp-Neo-Bam •lipofect AMINE™ alone '
Min/+ WT Min/+ WT
Day 1 2 1 1 1
Day 2 2 1 1 1
Day 3 2 1 1 1
Day 4 2 1 1 1
A total of three mice treated with APC pCMVp-Neo-Bam and two treated with
'LIPOFECTAMINE^'^ alone' were killed each day.
Table 2-Intra-peritoneal Trial
Treated with Treated with
APC pCMVp-Neo-Bam ’LIPOFECTAMINE™ alone'
WT WT
Day 1 2 1
Day 2 2 1
Day 3 2 1
Day 4 2 1
134 Figure 10- Agarose Gel Electrophoresis of rtPCR Products From Recto-
Sigmoid Tissue of All Rectally Treated Mice
Rectal Samples with Primers PI and P2
1396 bp
12345 12345 12345 12345 67 PP
Day 1 Day 2 Day 3 Day 4
Rectal Samples with Primers MBAF and MBAR
.mmrnm ë # # W w 872 bp
12345 12345 12345 12345 67
Day 1 Day 2 Day 3 Day 4
Samples 1 & 2 Min/+ mice treated with lipofectamine plus plasmid
Sample 3 WT mouse treated with lipofectamine plus plasmid
Sample 4 Min/+ mouse treated with lipofectamine only
Sample 5 WT mouse treated with lipofectamine only
Sample 6 Negative control. No RNA
Sample 7 Negative control. No reverse transcriptase
P Positive control, amplified plasmid cDNA
The first and last lanes on each gel are DNA ladders, of known molecular weight (1Kb ladder in the upper gel; 1 OObp ladder in the lower gel)
135 Figure 11- Agarose Gel Electrophoresis of rtPCR Products From
Colonic Tissue of Rectally Treated Mice Which Expressed The
Transgene In The Rectum
1 2 3 4 5 6 7 8 9 10 11 12 13 P P
1500bp 1396 bp 500bp 796 bp
872bp
1 2 3 4 5 6 7 8 9 10 II 12 13
Lane 1 lOObp DNA ladder
Lanes 2 to II Colonic samples from the ten mice (Min/+ and WT) which
expressed the transgene in their rectal tissue
Lane 12 Negative control. No RNA
Lane 13 Negative control. No reverse transcriptase
P Positive control, amplified plasmid cDNA
The samples in the upper row are amplified with Primers PI and P2. The samples in the lower row are amplified with primers MBAF and MBAR.
136 Figure 12- Agarose Gel Electrophoresis of rtPCR Products From The
Livers of Mice Treated With Rectal APC pCMVp-Neo-Bam
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 P P
ISOObp 1396bp
872bp
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Lanes 1 and 16 lOObp DNA ladder
Lanes 2 to 13 Colonic samples from the twelve mice (Min/+ and WT) treated
with APC pCMV(3-Neo-Bam
Lane 14 Negative control, No RNA
Lane 15 Negative control, No reverse transcriptase
P Positive control, amplified plasmid cDNA
The samples in the upper row are amplified with Primers PI and P2. The samples in the lower row are amplified with primers MBAF and MBAR.
137 Figure 13- Transgene Expression Following Rectal Treatment With
APC pCMVp-Neo-Bam
Daily Expression of Transgene
Number of
Animals
Expressing ® Rectum the ® Colon 3 Liver Transgene ^ Gonads
Dav2 Dav4
Overall Expression of Transgene
100
90 H Rectum Overall % 80 70 of Animals 60 Expressing 50
the 40
Transgene 30
20
10
0 Colon, Liver and Gonads
138 Figure 14- Agarose Gel Electrophoresis of rtPCR Products From The
Stomach of All Gavage Treated Mice
1 2345 12345 12345 12345 67 PP
1396bp
PI and P2
MBAF and
MBAR
12345 12345 12345 12345 67
Dayl Day 2 Day 3 Day 4
Samples 1 & 2 Min/+ miee treated with lipofectamine plus plasmid
Sample 3 WT mouse treated with lipofectamine plus plasmid
Sample 4 Min/+ mouse treated with lipofectamine only
Sample 5 WT mouse treated with lipofectamine only
Sample 6 Negative control, No RNA
Sample 7 Negative control No reverse transcriptase
P Positive control, amplified plasmid cDNA
The first and last lanes on each gel are lOObp markers.
139 Figure 15- Agarose Gel Electrophoresis of rtPCR Products From The
Proximal Small Bowel of All Gavage Treated Mice
12345 12345 12345 12345 67 PP
1396bp
PI and P2
MBAF and
MBAR
1 2345 12345 12345 12345 67
Y Dayl Day 2 Day 3 Day 4
Samples 1 & 2 Min/+ mice treated with lipofectamine plus plasmid
Sample 3 WT mouse treated with lipofectamine plus plasmid
Sample 4 Min/+ mouse treated with lipofectamine only
Sample 5 WT mouse treated with lipofectamine only
Sample 6 Negative control. No RNA
Sample 7 Negative control. No reverse transcriptase
P Positive control, amplified plasmid cDNA
The first and last lanes on each gel are lOObp markers.
140 Figure 16- Agarose Gel Electrophoresis of rtPCR Products From The
Livers of All Gavage Treated Mice
12345 12345 12345 12345 67 PP
■1396bp
PI and P2 H500bp
MBAF and - 872bp MBAR
Day 2 Day 3
Samples 1 & 2 Min/+ mice treated with lipofectamine plus plasmid
Sample 3 WT mouse treated with lipofectamine plus plasmid
Sample 4 Min/+ mouse treated with lipofectamine only
Sample 5 WT mouse treated with lipofectamine only
Sample 6 Negative control. No RNA
Sample 7 Negative control, No reverse transcriptase
P Positive control, amplified plasmid cDNA
The first and last lanes on each gel are lOObp markers.
Notice the low molecular weight bands in all samples amplified with PI and P2 primers on day 3 and in samples 1 to 3 on day 4.
141 Figure 17- Agarose Gel Electrophoresis of rtPCR Products From The
Gonads of All Gavage Treated Mice
12345 12345 12345 12345 67 PP 1396bp PI and P2 500bp
MBAF and 872bp MBAR
Samples 1 & 2 Min/+ mice treated with lipofectamine plus plasmid
Sample 3 WT mouse treated with lipofectamine plus plasmid
Sample 4 Min/4- mouse treated with lipofectamine only
Sample 5 WT mouse treated with lipofectamine only
Sample 6 Negative control, No RNA
Sample 7 Negative control. No reverse transcriptase
P Positive control, amplified plasmid cDNA
The first and last lanes on each gel are lOObp markers.
Notice the low molecular weight bands in many of the samples amplified with PI and P2.
142 Figure 18- Transgene Expression Following Gavage With APC pCMVp-
Neo-Bam
Daily Expression of Transgene
Number of 3
Animals ^ Stomach Expressing ■ Small Bowel 1 H Liver
^ Gonads Transgene
Dayl Day2 Day3 Day4
Overall Expression of Transgene
100
90
80 Overall % 70
of Animals 60
Expressing 50
the 40 Stomach 30 Transgene Small Bowel 1 20
10
0 HE Liver and Gonads
143 Figure 19- Agarose Gel Electrophoresis of rtPCR Products From
Peritoneal and Mesenteric Samples of All Mice treated via The
Intra-peritoneal Route
Peritoneal Samples
123 123 123 123 45PP 1396bp
PI and P2
MBAF and
MBAR 123 123 123 12345
Day 1 Day 2 Day 3 Day 4
Mesenteric Samples
123 123 123 12 3 4 5 P P
PI and P2
MBAF and
MBAR 123 123 123 1 2 3 4 5
Day 1 Day 2 Day 3 Day 4
Samples 1 & 2 WT mouse treated with lipofectamine plus plasmid
Sample 3 WT mouse treated with lipofectamine only
Sample 4 Negative control, No RNA
Sample 5 Negative control, No reverse transcriptase
P Positive control, amplified plasmid cDNA
The first lane on each gel is a lOObp DNA ladder, with bright bands at 500bp and 1500bp
144 Figure 20- Transgene Expression Following Intraperitoneal Treatment
With APC pCMVp-Neo-Bam
Daily Expression of Transgene
Number of ■ Peritoneum S Mesentery Animals 83 Liver ■ Small Bowel 1 Expressing @ Colon S3 Rectum the ^ Gonads
T ransgene D ay 1 D a y 2 D ay 3 D ay 4
Overall Expression of Trans gene
100 00 Peritoneum
90 1—fVh Mesentery and Liver 80 Overall % F4 Colon 70 of Animals HS3 Small Bowel 1 and Rectum 60
Expressing 50
the 40 H Gonads
Transgene 30 20
10
0
145 Table 3- Sites of Transgene Expression In Individual Animals After
Intraperitoneal Treatment With APC pCMVp-Neo-Bam
Peritoneum Mesentery Liver Gonads Small Colon Rectum
Bowel
Day 1 WTl Yes Yes Yes Yes
WT2 Yes Yes Yes Yes Yes
Day 2 WT4 Yes Yes Yes Yes Yes
WT5 Yes Yes Yes Yes Yes Yes Yes
Day 3 WT7 Yes Yes Yes Yes Yes
WT8 Yes Yes Yes Yes Yes Yes
Day 4 WTIO Yes Yes Yes Yes Yes
W T ll Yes Yes Yes Yes
146 Chapter 5
Clinical Effects ofAPC Gene Therapy
147 5.1 Materials and Methods
The clinical state of all the mice was recorded immediately after treatment and daily until death. The polyp load of Min/+ mice at the time of death was also recorded in the rectal and gavage trials. The intra-peritoneal gene therapy trial only included WT mice that did not develop polyps.
The position of each polyp was recorded as being in one of the intestinal segments described in Section 4.1.1. Polyp size was taken as the mean of two diameters measured with digital callipers (Micron Sales, London, UK). This gave a value to the nearest 1mm.
All data was recorded in Microsoft EXCEL 5.0 and analysed using non-parametric tests
(Mann-Witney U test) in SPSS 7.5 for Windows.
5.2 Results
5.2.1 Rectal Trial
The mice seemed generally unperturbed by treatment. There was no observed lack of appetite, no reduction in mobility and no change in social interactions. Only one Min/+ mouse treated with 'LIPOFECTAMINE™ alone' became non-specifically unwell on Day 1 and was humanely killed. Another member of the same litter (that had not been included in any part of the experiment) displayed similar symptoms of increased sleepiness and hunched appearance and was also killed to avoid distress. A thorough post mortem was performed on both mice as described (Section 4.1.1). There was no obvious abnormality that could have been attributed to the trial. Specifically there was no evidence of rectal perforation, no free fluid in the abdomen, and all abdominal organs and the recto-sigmoid mucosa looked macroscopically normal.
148 Examination of the rectal mucosa following enema treatment did not reveal any macroscopic inflammatory change.
Considering all Min/+ mice in the rectal treatment trial, the overall number of tumours in the small bowel ranged from 11 to 16. This was lower than levels previously reported
(Moser et al., 1990, Arenas et al., 1996, Wasan et al., 1997), for reasons discussed below.
The polyp load seen in the recto-sigmoid and colonic samples was also small, not rising
above a total large bowel count of three tumours. The majority of the polyps were one to two millimetres in diameter. Fewer than 10% of small bowel polyps and no large bowel
polyps approached three millimetres in size.
No alteration in overall large bowel, recto-sigmoid or colonic polyp number was recorded
in those mice treated with rectal APC pCMVp-Neo-Bam. Similarly there was no
alteration in the polyp count in the six Min/+ recto-sigmoid samples shown to express the
transgene when compared to the six samples (two treated with plasmid and four treated
with ’lipofectamine ^'^ alone’ that did not express the transgene (see Figures 21 and
22).
No polyps were seen in any of the mice genotyped to be WT.
5.2.2 Gavage Trial
No mice treated by gavage suffered any noticeable ill effects. In particular there was no
observed lack of appetite, no reduction in mobility and no change in social interactions.
Examination of the gastric mucosa following gavage did not reveal any macroscopic
inflammatory change.
149 The overall polyp distribution seen in these Min/+ mice was very similar to that recorded
in the rectally treated group (see Figure 23). Once again there was no significant
alteration in overall small bowel tumour count or in the number of polyps recorded in any
individual section of small bowel in those mice treated with the plasmid when compared to those treated with ’LIPOFECTAMINE™ alone*. Initial tumour count in the most proximal small bowel samples revealed a general trend towards a lower polyp number in
the three mice expressing the transgene when compared with those that failed to express the exogenous mRNA. However, there was no statistically significant difference (see
Figure 24).
5.2.3 Intra-peritoneal Trial
Treatment of the mice with an intra-peritoneal injection was well tolerated in all cases.
Once again there was no evidence of an alteration in appetite, mobility or group
interactions. Examination of the peritoneum and serosal surfaces within the peritoneal
cavity did not reveal any macroscopic inflammatory change.
5.3 Discussion
5.3.1 Unwanted Systemic Effects of Treatment
The illness seen in one mouse following rectal treatment with 'LIPOFECTAMINE™ alone*
may have been coincidental, as none of the other animals treated at the same time
displayed these symptoms and a non-treated littermate became ill at the same time. In the
absence of any other recorded adverse effects, it appears that all three routes of
administration of APC pCMVp-Neo-Bam are well tolerated and safe. There was no
consistent adverse response to either the lipid or plasmid moieties.
150 5.3.2 Beneficial Effect of Treatment on Polyp Number
Most studies in Min/+ mice have considered adult animals and those coming towards the end of their natural life, when the polyp load is maximal with 35 to 45 small bowel polyps and two or three large bowel neoplasms (Moser et al., 1990, Arenas et al., 1996, Wasan et al., 1997). In contrast, the mice used in this study were very young, approximately four weeks of age and had not yet entered puberty. This stage was chosen for treatment based on the proposition that corrective gene therapy has the best chance of working when administered at the earliest stage.
It has also been noted previously that ICRF Min/+ colonies show a lesser polyp load than the mice originally described by Moser's group (Wasan et al., 1997, Moser et al., 1990).
This may be due to dietary factors (Wasan et al., 1997) and/or the effect of modifier genes in the inbred strains of C57BL/6J WT mice used to breed the Min/+ heterozygotes.
Aberrant crypt foci (ACF) are small areas of intestinal epithelium that show irregular glandular architecture but not dysplasia. They are thought to represent the earliest stages in adenoma formation and it is known that they contain APC mutations (Smith et al.,
1994). Therefore, it has been suggested that therapeutic interventions in Min/+ mice could be judged by their effect on ACF formation rather than polyp load, and that this would circumvent the problems of low polyp numbers. Unfortunately although ACF identification has been pioneered in some centres (Reitmair et al., 1996a) it is generally accepted that it is a technique that shows great inter and intra-observer variability and it was felt that in the absence of local expertise this was not a technique to be relied upon for consistent results.
151 5.3.2.1 Rectal Triai
Absence of an alteration in large bowel polyp load following enema treatment with APC pCMVp-Neo-Bam was not surprising. A study in 1996 by Arenas et al treated 10 Min/+ mice rectally with this same plasmid (though in smaller doses, lOpg, and smaller final treatment volumes, 0.2ml) every three days for two months (Arenas et al., 1996). Rectal expression of the exogenous mRNA was demonstrated but no subsequent alteration in polyp load was recorded. It was suggested that the low polyp density within the rectum and colon of the Min/+ mice was inadequate to reveal a clinically small but statistically significant change. The same problem is present here. An additional factor in this study was the relatively short exposure of the intestinal mucosa to the transgene, ranging from
24 hours to four days.
However, appreciating this problem at the beginning of the project, it was decided to consider other phenotypic end-points, including expression of the 'normal' human APC protein and measurement of intra-cellular P-catenin levels in the treated large bowel
mucosa (See Chapter 6 ).
5.3.2.2 Gavage Trial
At the initiation of this project no work had been done looking at APC gene transfer in to small bowel mucosa. However, in October 1998, During et al reported that orally administered P-galactosidase transgene (in an AAV vector) lead to persistent expression of the gene in both gut epithelial and lamina propria cells of treated rats, and was associated with phenotypic improvement in lactose tolerance (During et al., 1998). This suggested that small bowel was a good site for gene transfection (given an appropriate vector) and that sustained transgene expression was possible here.
152 In this project it was postulated that the tumour burden in the small bowel of the Min/+ mice was a better background to assess clinical response to APC gene therapy and that
Min/+ small intestinal mucosa was potentially more abnormal than colorectal mucosa so this was also a better tissue in which to study all phenotypic changes resulting from the administration of APC pCMVp Neo-Bam.
However, only three of the eight Min/+ mice treated orally with the plasmid expressed the transgene in small bowel mucosa. The possible reasons for this are discussed in Section
4.3.5.1. Although there appeared to be a trend towards a decrease in polyp load in these mice (two killed after 48 hours and one killed after 72 hours), this was not statistically significant. This may have been due to the relatively short exposure to the transgene but it is also difficult to infer anything from such a small number of animals. Other phenotypic
effects of APC gene therapy are considered in Chapter 6 .
Recent work published by Arenas’ group looked at oral administration of APC pCMVp-
Neo-Bam in Min/+ mice (Lew et al., 2000). The mice were treated with plasmid alone
(n=4), with a combination of the plasmid and a COX-2 inhibitor (n=4), or with the same
COX-2 inhibitor alone (n=4) and compared to a control group of 6 Min/+ mice.
Following two months of biweekly treatments with 20pg of plasmid (approximately half of the treatment dose administered here) the mice were killed and the tumour burden
measured. A dramatic reduction in distal small bowel polyp load (84%) was seen in
animals treated with the combination therapy, and APC gene therapy alone reduced the
tumour number in this segment of bowel by 53%. It is interesting to note that the greatest
effect was on the most distal segment of small intestine, when this project would suggest
that transgene expression was minimal beyond the proximal small bowel even in the presence of larger doses of plasmid DNA. Unfortunately the levels of transgene
153 expression were not reported in this abstract, neither was the correlation between transgene expression and polyp load.
A final confounding factor in the use of lipid vectors in upper gastro-intestinal gene therapy is the intrinsic effect of the lipid on the modulation of disease state. It has been suggested that lipids have a protective effect against hydrophobic bile salt induced cell membrane damage in the upper G.I tract (Sagawa et al., 1993). Lipids are also commonly used as vehicles for drug administration, to enhance cell membrane transfer. Thus it is possible that the increased effect of combined COX-2 inhibition and APC pCMVP-Neo-
Bam noted by Arenas et al was due to a cytoprotective effect of the liposome construct and/ or improved uptake of the COX-2 inhibitor into the enterocytes.
154 Figure 21-Polyp Count In The Intestine Of Individual Min/+ Mice
Treated With Recta! APC pCMVP-Neo-Bam or 'LIPOFECTAMINE TM
Alone’
Small bowel 1 Small bowel 1
Small bowel 2 Small bowel 2
N um ber Small bowel 3 Small bowel 3 o f M ice
Colon Colon
Rectum Rectum O o o o ♦ ♦ o o ♦ ♦
012345678 01234567 8
Number of Polyps
^ Treated with APC pCMVp-Neo-Bam (n=8)
O Rectal samples that expressed the transgene after treatment with APC pCMVp-
Neo-Bam (n=6)
^ Treated with 'LIPOFECTAMINE^'^ alone' (n=4)
55 Figure 22- Comparison of Median Polyp Load in The Large Intestine of
Min/+ Mice Treated With Rectal APC pCMVP-Neo-Bam or
'LIPOFECTAMINE™ Alone'
Complete Large Bowel 3.0
2.5 - Colon
2.0 -
Number 1.5 - Rectum Rectum of 1.0 H Polyps
0.5
0
N= 8 4 6 6
I I Min/+ Mice treated with APC pCMVp-Neo-Bam
E Min/+ Mice treated with 'LIPOFECTAMINE^^ alone'
I I Min/+ mice treated with APC pCMVp-Neo-Bam and expressing transgene
g Min/+ mice treated with APC pCMVp-Neo-Bam but not expressing the
transgene
— Median Number of polyps
The height of the box is the interquartile range, which contains 50% of the values.
The whiskers extend to the highest and lowest values. There were no outliers.
N= number of mice used for data collection
156 Figure 23-Polyp Count In The Intestine of Individual Min/+ Mice
Treated By Gavage With APC pCMVp-Neo-Bam or
'LIPOFECTAMINE™ Alone'
Small bowel Small bowel 1 ^ ♦ O ❖ ❖ OOP Small bowel 2 Small bowel 2 O ♦ ♦ ♦ ♦ ♦ Number o o ^ Small bowel 3 Ÿ Small bowel 3 ♦ of Mice
Colon Colon ♦ ♦ ♦ ♦ Rectum Rectum ♦ ♦ ♦ ♦
012345678 01234567
Number of Polyps
^ Treated with APC pCMVp-Neo-Bam (n=8)
O Small Bowel 1 samples that expressed the transgene after treatment with APC
pCMVp-Neo-Bam (n=3)
Treated with T ipofect AMINE'^'^ alone' (n=4)
157 Figure 24- Comparison of Median Polyp Load in The Small Intestine of
Min/+ Mice Treated By Gavage With APC pCMVp-Neo-Bam or
'LIPOFECTAMINE™ Alone'
X 15 - I I ! Number
of 10 - Small Small Small Small Complete Polyps Bowel Bowel Bowel Bowel Small 1 1 2 3 Bowel
5 - H f Î I
0 N= 8 4 3 9 8 4 8 4 8 4
□ Min/+ Mice treated with APC pCMVp-Neo-Bam Min/+ Mice treated with 'LIPOFECTAMINE"^ alone' □ Min/+ mice treated with APC pCMVP-Neo-Bam and expressing transgene Min/+ mice treated with APC pCMVp-Neo-Bam but not expressing the
transgene
— Median Number of polyps
The height of the box is the interquartile range, which contains 50% of the values.
The whiskers extend to the highest and lowest values. There were no outliers.
N= number of mice used for data collection
158 Chapter 6
Protein Alterations Oçcurring After APC Gene Therapy
159 6.1 Introduction
It is postulated that successful APC gene therapy leads to the production of additional copies of the hill-length human APC protein with demonstrable downstream molecular effects. This chapter describes attempts made to identify the murine Ape protein and to quantitate p-catenin in the intestinal epithelium of untreated Min/+ and WT animals.
Following treatment with APC pCMVp-Neo-Bam, alterations in the expression of these proteins were recorded.
TSlormal' tissues in the Min/+ mouse are heterozygous for WT and truncated Ape proteins.
Therefore, immunoblot analysis of Ape was expected to reveal a 300kD band corresponding to the full-length protein and an 85kD band for the mutant form.
Translation of a functional protein as a result o f APC gene therapy was expected to cause an increase in the level of the 300kD band relative to its low molecular weight counterpart. (It is not possible to simply differentiate human APC from murine Ape due to their high level of homology at both the nucleotide and amino-acid level (Su et al.,
1992)).
6.2 Materials and Methods
6.2.1 Western Blotting
Western blotting (or immunoblotting) is a useful technique where electrophoretically
separated proteins are transferred from a gel to a solid support and probed for particular
sequences of amino-acids (Burnette, 1981). In this way specific proteins within complex mixtures can be identified and quantitated without radiolabelling. The procedure can be roughly divided into eight stages-
i) Preparation of the protein sample
Proteins were extracted from frozen tissue samples using a variety of methods 160 ii) Quantitation of protein concentration
This ensured equal loading of samples onto the gel iii) Denaturing gel electrophoresis of the protein sample
Proteins in each tissue lysate were separated according to molecular weight iv) Transfer to a membrane support
The protein network was given structural support for further intervention v) Blocking non-specific antibody (Ah) binding sites on the membrane
Using a mixture of proteins found in milk vi) Addition of specific primary and secondary Ab
A primary Ab, specific for the protein of interest, was added followed by a secondary horseradish peroxidase (HRPO) labelled Ab that specifically recognised the constant fragment of the primary Ab vii) Detection
A chemiluminescent signal was generated by the addition of a HRPO substrate. The resultant image was captured by autoradiography viii) Densitometry and Statistical Analysis of Results
To ensure that the initial protein electrophoresis had worked, the first gels produced with each method were stained to visualise the distribution of protein bands along the gel.
Following transfer of proteins to nitrocellulose membranes the first membranes were
similarly stained to ensure that protein transfer had proceeded with no complications.
These staining techniques are also described.
6.2.1.1 Preparation of the Protein Sample.
The aim of tissue lysis and protein extraction is to obtain all the protein of interest from the appropriate cellular compartment with minimal proteolysis. Frozen tissue samples
161 from the mid small bowel and colon were weighed and then homogenised for 2 0 minutes
in 1 0 volumes of the appropriate lysis buffer using a hand-held 2 ml mortar tube homogeniser (Merck Ltd, Poole, Dorset, UK). In order to minimise proteolysis all procedures were carried out on ice and pro tease inhibitors were present in all solutions
(Complete™ Protease Inhibitor Cocktail Tablets, Boehringer Mannheim, GmbH,
Mannheim, Germany, see Appendix 6 ). Once tissue lysis was complete, each sample was centrifuged at 13,000g and 4°C for 10 minutes to remove cellular debris and leave a cleared protein lysate. Lysates were either used immediately or stored in aliquots at -80®C until required.
6.2.1.1.1 Ape
A variety of previously described lysis buffers were used in attempts to extract the large
Ape protein, concentrating on two buffers described in Appendix 6 (Smith et al., 1993,
Chop et al., 1995). Tissue lysis was also attempted by boiling samples for ten minutes in
these same buffers.
6.2.1.1.2 p-Catenin
Total' cellular p-catenin was obtained using a lysis buffer described by Alman et al
(Alman et al., 1997) (Appendix 6 ). Following tissue homogenisation on ice, each sample was heated for ten seconds (in a microwave on high power) prior to centrifugation as
described above.
6.2.1.2 Quantitation of Protein Concentration
The amount of total protein in each tissue lysate was assayed so that equal amounts of total protein could be loaded onto each well of the gels. A bicinchoninic acid (BCA)
162 colourimetric assay was used (BCA Protein Assay Kit, Pierce, Rockford, Illinois, USA).
The dye BCA undergoes a differential colour change in response to various concentrations of protein (Smith et al., 1985). The depth of colour of the solution is measured by its absorbance of light at 562nm wavelength. The reaction is described in
detail in Appendix 6 .
6.2.1.2.1 Protocol
• Samples were assayed at full concentration and two, five, ten and twenty fold
dilutions in lysis buffer.
• Standards were known concentrations of bovine serum albumin, increasing from 0 to
lOOOpg/ml in steps of 200pg/ml, diluted in the same lysis buffer as the samples.
• 50pl of each standard or unknown protein sample was pipetted into an appropriately
labelled eppendorf tube.
• BCA reagents A and B (Appendix 6 ) were mixed together in the ratio 50:1 by
volume. One millilitre of the resultant solution was added to each tube and incubated
for 30 minutes at 37®C.
• OD of standards at 562nm incident light was measured and used to construct a
standard curve of OD against protein concentration using a least squares regression
model (Microsoft Excel version 5.0) as shown in Figure 25.
• From the formula of the curve, the OD at 562nm of each sample could be extrapolated
to give an estimation of its protein concentration.
• All measurements were taken within a ten minute period with the tubes kept on ice
until the samples were poured into the cuvette. This slowed any ongoing reaction and
prevented the colour from deepening.
163 6.2.1.3 Denaturing Gel Electrophoresis
Proteins are separated according to molecular weight by their rate of movement through a polyacrylamide gel matrix in the presence of an electrical gradient. The percentage of the acrylamide used is varied in line with the protein under consideration. High percentage gels (with small pore size) give good resolution of small proteins and vice versa.
Protein preparation and electrophoresis is performed under conditions that maintain dissociation of proteins into their individual peptide subunits. For example, the p-catenin protein preparations were boiled in loading buffer containing P-mercaptoethano 1 and SDS prior to loading onto the gel. p-mercaptoethanol retards oxidation of biological compounds in solution and SDS binds to peptides. The amount of SDS bound is directly proportional to the molecular weight of each peptide and its strong anionic charge negates any polarity effect related to the peptides.
The proteins run first through a stacking and then through a resolving gel in a discontinuous buffer system devised by Ornstein and Davis (Omstein, 1964, Davis,
1964). The running buffer is a different pH and ionic strength to the stacking gel.
Therefore, when an electric current is passed between the electrodes, the SDS polypeptide complexes are quickly swept through the stack and deposited on the top of the resolving gel in a very thin zone (or stack). A higher pH within the resolving gel then allows the
SDS-protein complexes to move through the gel at a rate proportional to their size.
Coloured marker proteins of known molecular weight (Sigma prestained SDS-PAGE standard solution. Sigma-Aldrich Company Ltd or Rainbow^^ coloured protein molecular weight markers, Amersham Pharmacia Biotech Ltd) were loaded onto the outermost wells of each gel to allow estimation of size of any positive band (Laemmli, 1970).
164 6.2.1.3.1 Apc
Although SDS-polyacrylamide (SDS-PAGE) is the normal matrix used to resolve a protein mixture, many experimenters have encountered difficulty in getting the large Ape protein (310kD) to enter the small pores of an SDS-PAGE gel. SDS-agarose gel electrophoresis was initially described in 1993 for the separation of high molecular weight proteins of the neurochordin family (Preobrazhensky, 1993) and has been used with some success in the study of the APC gene product (Smith et al., 1993). Therefore,
Apc protein samples were run out on thin horizontal SDS agarose gels in line with this
protocol (Appendix 7). Gels were run at < 5V/cm potential difference for over 8 hours, occasionally overnight at 4®C, until the low molecular weight markers had run off the end of the gel. All experiments were run in duplicate.
6.2.1.3.2 p-Catenin
Standard vertical SDS-PAGE gels were used, with an 8 % resolving gel (giving good resolution of proteins around the 90kD range), and a 3% stacking gel.
6.2.1.3.2.1 Protocol
• The glass plates were cleaned with soap and water, followed by ethanol and arranged
vertically around appropriate spacers and sealed with a clamp. All experiments were
run in duplicate with two gels placed back to back and loaded with identical samples.
• The resolving gel was prepared (Appendix 7) and poured between the plates up to a
level 1.5cm below the bottom of the teeth of the comb that defines the wells. A layer
of ethanol was poured on top of the gel to level off the upper surface and enhance
polymerisation. (Oxygenation slows polymerisation).
165 • 30 minutes later, the layer of ethanol was poured off, and the top of the gel washed
with ddH 2 0 to remove loose flakes of polyacrylamide.
• The stacking gel (Appendix 7) was prepared and poured on top of the resolving gel. A
comb was inserted to create the wells. The gel was left to set for a further 30 minutes.
• The comb was removed and the wells were irrigated with running buffer to remove
any loose flakes of polyacrylamide.
• The gels were either submerged in running buffer (Appendix 7) in an ATTO gel tank
(ATTO Dual Mini Slab Kit AE6400, GRI Ltd, Braintree, Essex, England) or the base
and top of the gel were positioned in two pools of buffer in a Hoefer SE 400
electrophoresis unit (Hoefer Scientific Instruments, San Francisco, California, USA).
The ATTO system was quicker to set up and run if the maximum number of wells
per gel was 12. However in larger experiments the Hoefer gel system held gels of
18x16cm in size, allowing upto 2 0 lanes per gel.
• 5pg of total protein from each sample was loaded into the wells. Coloured marker
proteins were placed in the outermost wells. An equal volume of loading buffer was
placed in unused wells. Electrophoresis was commenced at a constant current of
20mA per gel. This setting was chosen to avoid heating of the plates.
• When the dye front reached the bottom of the gel the power was turned off and the
gels removed from the plates. The gel was then used for staining or transfer.
6.2.1.4 Staining of Protein Gels
Gels can be stained for protein using several substances including Coomassie Brilliant
Blue R250 (a triphenylmethane dye) and Silver Salts (Appendix 8 ). Coomassie staining is the most commonly used technique (Sambrook et al., 1989c)
166 6.2.1.4.1 Coomassie Brilliant Blue Staining Protocol
• The gel was placed flat in a glass dish covered with at least five volumes of staining
solution (Appendix 8 ). The solution contains a mixture of methanol and glacial acetic
acid that 'fixes' the proteins within the gel.
• It was left on a rotating platform at room temperature for 1 to 4 hours (thin PAGE
gels stain more quickly than thicker agarose gels).
• Destaining was performed by tipping off this solution and soaking the gel in wash
solution (Appendix 8 ). This solution was renewed every hour and a piece of sponge or
tissue paper was placed in one comer of the dish to increase the resorption of the dye.
• The optimal time for destaining was when the background colour in the gel had all
been removed leaving clearly stained bands of protein. This took approximately 6
hours for PAGE gels but upto 48 hours for agarose gels. Even after this length of time
the bands were not as sharply defined.
• PAGE gels were placed onto a piece of Whatmann 3MM paper and covered with
Saran Wrap. This sandwich was placed in a heated (60®C) vacuum gel drier for 2
hours. The dry gels were stored. Agarose gels did not dry well due to their increased
thickness so they were photographed and discarded.
6.2.1.4.2 Silver Staining
Silver staining is at least 100 times more sensitive in protein detection than Coomassie
staining. It reveals nanogram quantities of protein. This was used in an attempt to detect the high molecular weight (310kD) Ape protein in denaturing agarose gels.
Silver nitrate (AgNOs) solutions were used based on a modification of the method of
Sammons et al (Sammons et al., 1981). The basic principle is the reduction o f silver
167 nitrate (AgNOs) to metallic silver at a protein band, leading to the deposition of silver
grains. The gel was first immersed in AgNOs at acidic pH. Once the AgNOs reacted with
protein sites, the ionic silver was reduced to metallic silver by formaldehyde at alkaline pH. Gloves were worn at all times as fingerprints produce staining artefacts. The gel was
placed flat in a clean glass dish at room temperature on a rotating platform.
6.2.1.4.2.1 Protocol
• Five gel volumes of fixative were used (Appendix 8 ) to fix the protein position within
the gel and to remove detergents and reducing agents which interfere with the staining
procedure.
• After four hours, the fixative was replaced with five gel volumes of 30% ethanol. This
step was repeated after a further 30 minutes.
• The ethanol was discarded and replaced with ten gel volumes of ddH 2 0 . The gel was
incubated for ten minutes and this step was repeated twice more.
• The last ddHzO wash was discarded and five gel volumes of aqueous 0.1% AgNOs
solution made up fi*eshly fi*om granules, was added. The gel was incubated in this
solution for 30 minutes.
• The AgNOg solution was poured off and the gel was carefully washed under running
ddHiO. The gel was not allowed to dry out as this can cause staining artefacts.
• Five gel volumes of a fi-esh aqueous solution of alkaline 2.5% sodium carbonate,
0 .0 2 % formaldehyde was added and the gel was incubated with gentle agitation until
stained protein bands appeared. This took one to three minutes. (Formaldehyde was
obtained as a 37% aqueous solution, which was diluted in a fume cupboard to prevent
vapour exposure).
168 • The reaction was quenched by washing the gel in 1% aqueous acetic acid for five
minutes and then repeated washing in ddHiO. The gel was photographed on a light
box at this point.
6.2.1.5 Transfer To A Membrane Support
Proteins are transferred from the gel to a nitrocellulose support by capillary transfer or direct electrophoretic transfer, the latter being the most commonly used method (Towbin et al., 1979, Burnette, 1981). The proteins become covalently bound to the membrane. All gels and membranes are handled wearing gloves as oils on the skin can impede protein transfer.
6.2.1.5.1 Ape
Ape protein lysates were transferred by capillary action, based on a protocol described for the transfer of nucleic acids to a membrane support after electrophoresis (Sambrook et al.,
1989a). This is not as efficient as electroblotting but agarose gels do not stand up well to electroblotting.
6.2.1.5.1.1 Protocol
• The top right hand comer of each gel was marked with a small cut. The gels were then
placed in a tray of transfer buffer (Appendix 9) for 10 minutes to equilibrate with the
buffer.
• A piece of nitrocellulose membrane and five pieces of Whatmann filter paper were
cut to the same size as the gel. Two strips of filter paper the same width as the gel but
several times as long were also prepared.
169 The membrane, three gel sized pieces of Whatmann paper and the long strips of filter paper were pre-soaked in a tray of transfer buffer.
Two pieces of dry Whatmann paper were placed on a two inch stack of blotting paper
(or paper towels) on a board above a tray of transfer buffer as shown in Appendix 9.
A pre-soaked piece of filter paper was placed on top of the pile. The sides of the blotting paper were wrapped in Saran wrap to keep them as dry as possible.
The pre-soaked membrane was placed on top of the damp filter paper. Care was taken to avoid trapping air bubbles that impede smooth capillary transfer.
The gel was placed on top of the membrane. Once again it was important to avoid trapping air bubbles.
The two remaining pre-soaked pieces of Whatmann paper were placed on top of the gel and covered by the long strips of filter paper. Both ends of these strips communicated with the transfer buffer.
The entire apparatus was covered in Saran Wrap and a heavy weight was placed on top.
The transfer was left to happen overnight. A flow of buffer was set up by capillary action fi-om the tray through the gel and the membrane to the dry Whatmann paper and blotting paper. Protein was therefore eluted fi*om the gel and became covalently bound to the nitrocellulose membrane.
The next day the membrane was removed for staining or Western blotting. The appearance of the coloured markers on the membrane indicated that protein transfer had occurred.
170 6.2.1.5.2 P-catenin
Protein samples run on SDS-PAGE gels were transferred to nitrocellulose membranes using conventional wet electroblotting techniques (Harlow and Lane, 1988) in a Bio-Rad
Trans-blot Cell ( Bio-Rad Laboratories Ltd).
6.2.1.5.2.1 Protocol
• Three litres of transfer buffer were prepared (Appendix 9) and chilled to 4®C.
• The gels were removed from the electrophoresis apparatus, the stacking gels were
discarded and the resolving gels were placed in cold transfer buffer to equilibrate for
ten minutes. The top right hand comer of each resolving gel was marked with a small
cut.
• Two pieces of Whatmann paper and a piece of nitrocellulose membrane were cut to
the size of the resolving gel. They were pre-soaked in transfer buffer along with two
porous pads (part of the transfer equipment).
• One gel and a sheet of nitrocellulose membrane were sandwiched between the pre-
soaked pads and Whatmann paper (as shown in Appendix 9) and submerged in
transfer buffer in the electrophoresis tank. Care was taken to avoid the inclusion of air
bubbles that inhibit protein transfer.
• The tank contained standard platinum electrodes mounted on either side wall.
• The nitrocellulose membrane was placed towards the anode.
• A small magnetic bar was placed in the bottom of the tank and the whole apparatus
was placed on top of a magnetic stirrer to prevent overheating and the consequent
formation of air bubbles in the sandwich.
• A constant electric current of 100mA was applied overnight.
171 • The negatively charged proteins migrated from the gel to’svard the anode and became
covalently attached to the nitrocellulose membrane.
6.2.1.6 Staining of Proteins Covalently Bound To The Nitrocellulose Membrane
Many techniques have been described. In this project, staining with Ponceau S was routinely used to estimate protein transfer as the stain is completely removed during washing so it does not interfere with subsequent immunoblotting (Sambrook et al.,
1989b). Unfortunately the pink colour of the stain is difficult to capture photographically
so there is no permanent record of the transfer. It merely provides visual evidence of transfer and allows the position of molecular weight markers to be drawn on the membrane. Staining with India Ink (a more sensitive but permanent stain) was
occasionally used to provide a permanent record of protein transfer.
6.2.1.6.1 Ponceau S Staining Protocol
• The top right hand comer of each membrane was marked in pencil and it was placed
in a tray o f ddH 2 0 on a rotating platform at room temperature for five minutes.
• The membrane was placed in a working solution of Ponceau S stain for five to ten
minutes (Appendix 10).
• The membrane was placed back in ddHiO. The protein bands gradually became more
obvious as the background colour was eluted.
• Protein markers were labelled where necessary and washing continued for 15 minutes
with frequent changes of ddIÎ 2 0 .
• Immunological probing could be performed straight away.
172 6.2.1.6.2 India Ink Staining Protocol
• The top right hand comer of each membrane was marked in pencil and it was placed
in a tray of ddHiO on a rotating platform at room temperature for five minutes.
• The membrane was placed in soaking solution (Appendix 10) for ten minutes,
renewing the solution after five minutes. Tween 20 (Sigma Aldrich Company Ltd) is a
detergent in the solution that acts to reduce non-specific binding of colloidal particles
(in the Ink) to the membrane.
• The staining solution (Appendix 10) was added and the filter was left to incubate until
protein bands appeared. This took a variable length of time. Most protein bands
appeared after ten to twenty minutes but when looking for Ape (heavier then the
largest protein marker) the membranes were left to stain for upto several hours.
• The membrane was washed several times in PBS to clear the background staining and
photographed.
6.2.1.7 Blocking Non-Specific Binding Sites On The Membrane
Nitrocellulose membranes avidly bind all proteins. In order to minimise non-specific binding of Ab, the membrane was exposed to a blocking solution of irrelevant proteins.
Several blocking systems have been described but the cheapest and easiest to use is non fat dried milk (Marvel^^) first described by Johnson et al (Johnson et al., 1984). The membrane was 'blocked' prior to antibody additions and all antibody solutions were made up in blocking solution.
173 6.2.1.7.1 Protocol
• The nitrocellulose membrane was washed briefly in TBST (Appendix 11). The top
right hand comer of each membrane and the position of marker proteins were labelled
in pencil.
• The membrane was then placed in blocking solution (Appendix 11) on a rotating
platform at room temperature for one hour. One millilitre of solution per cm^ of
membrane was used.
6.2.1.8 Addition Of Primary and Secondary Antibodies
A primary Ab, specific for the protein of interest, was added. Although it is possible to directly label the primary Ab, an indirect (or secondary) labelling system was used here to facilitate the use of multiple different primaries.
The primary Ab binds a specific epitope on the target protein and the secondary Ab, which is labelled for easy detection, binds specifically to the Fc portion of the primary.
The primary Abs used were all monoclonal mouse IgGl. Various Ab dilutions were tested to find the concentration that gave the cleanest signal at a reasonable exposure time
(from 20 seconds to 1 hour).
6.2.1.8.1 Protocol
• The primary Ab was made up at the appropriate concentration in blocking solution.
• One hundred micro litres of Ab solution/cm^ of membrane was placed in a heat-
sealable plastic bag with the pre-blocked membrane. All bubbles were removed and
the bag was sealed close to the edge of the membrane.
• The bag was placed fiat on a rotating table at room temperature for two hours.
174 • The membrane was removed from the bag and washed in TEST (Appendix 11) for 30
minutes, renewing the wash solution every ten minutes. This washing step aimed to
remove excess Ab bound non-specifically to the membrane.
• The secondary Ab, a high specificity HRPO conjugated sheep anti-mouse Ig (NA 931,
Amersham Pharmacia Biotech Ltd.) was made up at 1/5000 dilution in blocking
solution.
• One hundred micro litres of this solution/cm^ of membrane was placed in a heat-
sealable plastic bag with the washed membrane. All bubbles were removed and the
bag was sealed close to the edge of the membrane.
• The bag was placed flat on a rotating table at room temperature for one hour.
• The membrane was removed and washed in TEST for one and a half hours, changing
the wash solution every fifteen minutes. This final extensive wash procedure removed
excess non-specifically bound secondary Ab.
6.2.1.9 Detection of Signal
Detection of Ab bound specifically to nitrocellulose membrane is based on the principle of enhanced chemiluminescence (Whitehead et al., 1979). Under alkaline conditions, and
in the presence of hydrogen peroxide (H 2 O2 ), HRPO (conjugated to the Fc portion of the secondary Ab) catalyses the oxidation of a cyclic diacylhydrazide (luminol). Oxidised luminol is in an excited form and it decays with the production of light (luminescence), as shown overleaf. In the presence of an oxidation enhancer (such as phenol) the light output is both increased and prolonged. The wavelength of the emitted light (462nm) affects radiographic film (autoradiography). The response of the film is proportional to the amount of protein present as long as the primary and secondary Abs and the chemiluminescence solutions are not limiting.
175 HRPO-Ab Protein
Oxidised luminol luminol
Nitrocellulose Light at 462nm membrane
Throughout this project ECL*”^ Western blotting detection reagents and Hyperfilm"^ ECL
(Amersham Pharmacia Biotech Ltd.) were used according to the manufacturer's instructions.
6.2.1.9.1 Protocol
• Excess TBST was drained from the membrane, and it was placed in a flat plastic tray
with the surface of the membrane that had been directly opposed to the gel during
protein transfer uppermost.
• ECL^”^ reagents A and B were mixed in equal proportions to give a final volume of
0.125ml/cm^ of membrane. This mixture was poured directly onto the membrane (it
contained H 2 O2 , luminol and chemiluminescence enhancers in an alkaline medium).
• After one minute the membrane was removed from the tray and washed briefly in
ddEl20. Excess ddH20 was drained from the membrane, it was wrapped in Saran
wrap, placed protein side up in an autoradiography cassette and transferred to a dark
room. 176 • Hyperfilm™ ECL was exposed to the membrane. Exposures varied in length from
five seconds to one hour, trying to visualise the weakest signal on each membrane
without overexposing too many bands. In the presence of a very weak signal the film
was left to expose overnight.
6.2.1.9.2 Signal Calibration
The initial protein loading onto the gel was adjusted until immunolabelling gave a visible but not excessive signal at a reasonable exposure time (minutes to hours). The concentrations of primary and secondary Abs and the volume of wash solution were adjusted to achieve a clear signal with minimal background.
6.2.1.10 Stripping and Re-probing Membrane
Membranes that had been probed for P-catenin were subsequently stripped of Ab and chemiluminescence reagents and re-probed for p-actin and/or cytokeratin. p-actin is a ubiquitous protein and equivalent signals from adjacent lanes act as a control for equal protein loading. Cytokeratins are proteins found only in epithelial tissues.
Immunoblotting for these proteins allowed standardisation of the amount of epithelium within the tissue extract.
6.2.1.10.1 Protocol
• Following completion of autoradiography the membrane was placed in a glass
container with 500ml of Strip Buffer (Appendix 11) and left to incubate overnight at
room temperature on a rotating platform.
• The membrane was washed in TBST for thirty minutes, changing the solution every
ten minutes.
177 • The immunolabelling procedure was repeated as described above, using a different
primary Ab.
• If blotting for another protein was to be delayed for any reason the membrane was
stored refrigerated in TBST until required.
In order to ensure the process worked well the initial membranes were exposed with
ECL™ chemiluminescence reagents after stripping and no signal was generated.
6.2.1.11 Densitometry and Statistical Analysis Of Results
The best autoradiograph from each Western blot was scanned on a flat bed scanner at a resolution of 1,000 dots per inch, using Adobe Photoshop version 4.1 software. The 'best' exposure was taken to be that where the maximum number of samples gave a visible signal to be quantitated (so that they could be compared to each other) with the minimum number of samples having a density off the bottom or top of the scale. All autoradiographs were scanned twice and the results were only accepted if they differed by
< 10%.
NIH Image version 1.62 was used to analyse signal density. The band density was recorded along with the background density from a point at the top of the same lane. The background was subtracted from the signal density to give the final measurement. Density
was recorded on a scale of 1 to 255 units.
All gels were run in duplicate so that two sets of results were generated for each
experiment and many experiments were performed at more than one time point. All data was entered into Microsoft Excel version 5.0 and subsequently analysed using non- parametric tests (Mann-Whitney U for two independent variables and Kruskal-Wallis
Test for multiple independent variables) in SPSS version 7.51.
178 6.2.2 Immunohistochemistry
Immunohistochemistry localises target proteins within tissue sections using a labelled Ab
directed towards a known antigen (Ag) on the protein of interest. It relies upon a strong and specific Ab/Ag interaction.
In this project immunohistochemistry was used to substantiate the results derived by
immunoblotting. Frozen tissue sections were used instead of paraffin fixed sections so that all tissue samples could be collected in the same way, snap frozen and stored at -
80®C. Primary Abs to Ape or p-catenin were used, described below. Indirect labelling of the primary Ab was used in order to generate a highly specific signal with minimal background. A Secondary Ab (specific for the Fc portion of the primary) was conjugated to biotin. Streptavidin-HRPO was then added. The streptavidin bound tightly to the
biotinylated Fc portion of the secondary Ab. Finally a solution containing the substrate
diaminobenzidene (DAB) was added. The HRPO converted DAB to a brown precipitate
that was visualised on the tissue sections by microscopy.
ImmunoCruz Staining System kits (SantaCruz Biotechnology Inc., Santa Cruz,
California, USA) were used. These contained ready diluted primary and secondary Abs
and all other necessary solutions. The primary Abs were anti-human, affinity purified,
polyclonal rabbit and goat Abs raised against the carboxy-terminus of APC and P-catenin
respectively. Both were known to react with mouse homologues of the human proteins.
Negative control sections were treated with 'normal' rabbit or goat IgG instead of primary
Ab but were otherwise treated in an identical fashion. One to three drops of pre-diluted
solutions were applied to each section dependant upon its size. All steps were carried out
at room temperature in a moist chamber unless otherwise stated.
179 6.2.2.1 Protocol
• Frozen tissue specimens from mouse mid small bowel or colon were mounted in
O.C.T. ™ compound in a cryostat maintained at -20®C (Merck, Poole, UK).
• Serial 5 pm sections were cut, collected onto Vectabond treated glass slides (Vector
Laboratories, Burlingame, USA), air-dried for two hours at room temperature and
stored at -80^C.
• Every fifth section was stained with haematoxylin and eosin (H&E) using standard
methods (Appendix 12) to define overall morphology and to check the quality of the
sections prior to immunohistochemical labelling.
• The remaining sections were thawed and then fixed in 100% acetone at 4®C for seven
minutes.
• Sections were washed in room temperature PBS for fifteen minutes, changing the
wash solution every five minutes.
• Endogenous tissue peroxidase activity was blocked by the addition of 4% H 2 O2 for
five minutes.
• The slides were washed in PBS for four minutes, changing the wash solution after two
minutes.
• A 5% serum block (from the same species as the secondary Ab) was applied for 20
minutes to prevent later non-specific sticking of the secondary Ab.
• Excess unbound serum was aspirated from the section.
• The slides were incubated with primary Ab for 2 hours.
• They were washed in PBS for four minutes, changing the wash solution after two
minutes.
• Biotinylated secondary Ab was applied for 30 minutes.
• The slides were washed in PBS for four minutes, changing the wash solution after two
minutes.
180 • Streptavidin-HRPO complex was added and the slides incubated for 30 minutes.
• The slides were washed in PBS for four minutes, changing the wash solution after two
minutes.
• HRPO substrate containing DAB was added. The colouration was allowed to proceed
for 10 minutes until visible with the naked eye.
• At this point the slides were washed in ddHzO, counterstained in Gill's haematoxylin
for five seconds, and washed thoroughly under running tap water.
• The sections were dehydrated by passage through a series of alcohols and xylene
using standard methods (as described in 'Haematoxylin and Eosin Staining of Tissue
Sections'-Appendix 12) prior to mounting in diphenylxylene (Merck, Poole, UK).
• Mounted sections were viewed using xl0(0.3NA) x20(0.5NA) and x40(0.75NA)
lenses of a Zeiss Axiophot microscope (Zeiss, Jana, Germany). Representative images
were captured using a Leaf Microlumina high resolution CCD colour camera (SS,
Greater Manchester, UK) controlled by proprietary software. The images were
collected together with an empty field of view that was used for background
correction of spatial shading using Image Pro-Plus software version 3.0.
6.3 Results
6.3.1 Untreated Animals
6.3.1.1 Western Blotting
Initial experiments to optimise experimental technique looked at intestinal samples fi'om
Min/+ mice and WT counterparts aged approximately 6 months, none of whom had received any kind of treatment. When the protocols were established, samples fi*om the mice in the rectal and gavage trials were analysed.
181 6.3.1.1.1 APC
Following quantitation of the Ape protein lysates, 'total protein' (from lOOpg to 400pg in incremental doses) was loaded onto SDS-agarose gels and the gels were run as described in Section 6.2.1.3.1. However, staining these gels with Coomassie Brilliant Blue did not reveal any protein band in the 300kD range. Subsequent experiments utilised silver staining, as this is more sensitive to protein, but once again there were no visible protein bands in the predicted size range. This was against a background of multiple clear bands at lower molecular weights. Repeating the experiments with SDS-PAGE gels of different concentrations did not alter these findings.
Immunoblotting routinely detects l-5ng of common proteins and as little as lOOpg of some proteins under ideal conditions. Therefore, it was decided to proceed with Western blotting in the hope that there would be sufficient Ape within the tissue lysates to allow immunolabelling despite there being too little to show up on protein stains. Unfortunately no clear specific signal was seen at any molecular weight even after overnight incubation of the membrane with the Hyperfilm ECL. This was despite initial loading of high protein concentrations and with the primary Ab diluted 1 in 100 instead of the 1 in 1000 dilution recommended by the manufacturer. The possible reasons for failure of this technique are discussed below.
The total lack of staining with Ab-1 meant that it was not possible to use alterations in the ratio of full length to truncated Ape as a marker for successful gene therapy.
182 6.3.1.1.2 p-Catenin
Figures 26 and 28 show representative Western Blots of P-catenin extracted from Min/H and WT recto-sigmoid and small bowel tissue lysates respectively. In each figure these autoradiographs are shown in conjunction with autoradiographs of the same membranes stripped and then probed for p-actin and cytokeratin. The monoclonal anti-pan cytokeratin Ab used in this project was chosen as it reacted with mouse cytokeratins and could be used in combination with the same secondary Ab as both the p-catenin and p- actin primary Abs. One difficulty in its use was that it produced multiple bands, corresponding to cytokeratins one, five, six and eight. The cytokeratin eight signal (a band at 52kD) came up most strongly in all experiments and was used for quantitation.
Cytokeratin eight is a major type II keratin present in simple epithelia, whereas cytokeratins one and six are expressed differently dependant on the level of epithelial differentiation and proliferation respectively, and cytokeratin five is the primary cytokeratin in stratified epithelia.
Figure 27 is a graphical comparison of the level of p-catenin in untreated Min/+ and WT recto-sigmoid. The p-catenin signal was initially corrected for the amount of total tissue present, by correcting for the P-actin signal. Correction for the cytokeratin eight signal then allowed for the amount of epithelial tissue present.
The results suggested that there was no statistically significant difference between the amount of P-catenin present in the recto-sigmoid epithelium of Min/+ mice when compared to their WT counterparts (p= 0.827).
Figure 29 is a graphical comparison of the level of P-catenin in untreated Min/+ and WT small bowel. Using cytokeratin as a control, these results showed a statistically significant increase in the level of p-catenin in Min/+ small bowel mucosa when compared to WT 183 small bowel (p= 0.034). Using p-actin as a control a similar trend was seen although this did not reach statistical significance (p=0.289).
6.3.1.2 Immunohistochemistry
Frozen sections were cut on a cryostat. Every fifth section was stained with haematoxylin and eosin. Representative samples are shown in Figure 30. It can be seen that the tissue architecture obtained in frozen section is less detailed and more susceptible to artefact than that routinely achieved in paraffin fixed sections.
6.3.1.2.1 Ape
Immunohistochemical staining of frozen sections of recto-sigmoid and proximal small bowel taken from Min/+ (n=4) and WT (n=4) mice gave patchy results but did not reveal any obvious difference between the two groups. Labelling was limited to the epithelium and there was no clear membranous or cytoplasmic preference (see Figure 31).
6.3.1.2.2 p-catenin
Labelling of p-Catenin in recto-sigmoid sections showed uniform labelling of cell membranes with a fainter cytoplasmic signal and negligible nuclear staining. It was impossible to differentiate between Min/+ and WT samples (see Figure 32a and b).
Immuno staining in small bowel sections revealed an increased level of staining in histologically normal Min/+ epithelium when compared to WT epithelium (see Figure
32c and d). The increased labelling was most noticeable within the cytoplasm and there was negligible nuclear staining.
184 6.3.2 Treated Animals
Following a trial of immunoblotting and immunohistochemistry to quantitate levels of
Ape and P-catenin in untreated animals, it was decided to focus on Western Blotting for
p-catenin as an outcome measure of gene therapy. Intestinal samples taken from the treated animals had been cut longitudinally into two. One half was used to assess transgene expression ( as described in chapter 4) and the remaining tissue was used for protein analysis using the technique outlined above.
6.3.2.1 Rectal Trial
Eighteen out of twenty recto-sigmoid samples were available for protein analysis. Two
samples (one from a Min/+ mouse treated with APC pCMVP-Neo-Bam that failed to express the transgene three days after treatment, and one from a Min/+ mouse treated
with 'LIPOFECTAMINE alone’ and sacrificed one day after treatment) were needed for repetition of inconclusive initial rtPCR experiments.
Figure 33 shows a typical Western Blot of the recto-sigmoid tissue lysates probed for p-
catenin, p-actin and cytokeratin.
Figure 34 is a boxplot representation of the level of P-catenin in these tissue samples
when grouped according to genotype and transgene expression. There was no significant
difference between the amount of P-catenin present in Min/+ recto-sigmoid samples not
expressing the transgene and WT recto-sigmoid samples not expressing the transgene.
This reflects the findings of section 6.3.1.1.2. However, transgene expression was
associated with a decrease in the median amount of p-catenin in mice of both genotypes,
whether corrected for p-actin or cytokeratin levels. This effect reached statistical
185 significance (p=0.019) in Min/+ mice when the result was corrected for the amount of epithelium present.
Figure 35 is a boxplot of the results derived from the same blot when samples were grouped according to treatment arm instead of transgene expression. This made little difference to the results as only one Min/+ mouse (treated with APC pCMVp-Neo-Bam but not expressing the transgene) changed group. There was no significant difference when the base-line level of recto-sigmoid P-catenin in Min/+ mice treated with
'LIPOFECTAMINE^"^ alone' was compared to that of WT mice treated in the same way.
Following treatment with APC pCMVp-Neo-Bam there was a consistent trend towards a decrease in the median value of p-catenin. When corrected for the amount of epithelium present this only just failed to reach statistical significance (p=0.053).
6.3.2.2 Gavage Trial
Fifteen out of twenty proximal small bowel samples were available for analysis, the
remaining samples having been needed for repetition of inconclusive rtPCR results. These were all mice which had been treated with'LIPOFECT AMINE^*^ alone' (1 Min/+ mouse
and 1 WT mouse) or who had failed to express the transgene in the proximal small bowel
after treatment with APC pCMVP-Neo-Bam (2 Min/+ mice and 1 WT mouse).
Figure 36 shows a typical Western Blot of the remaining fifteen proximal small bowel tissue lysates probed for p-catenin, p-actin and cytokeratin.
Figure 37 is a boxplot representation of the level of p-catenin in these tissue samples when grouped according to genotype and transgene expression. There was a consistently
186 significant difference between samples taken from Min/+ and WT mice that did not express the transgene. This was not affected by correction for total tissue content or epithelial content of the sample (p=0.004). Transgene expression in the proximal small bowel was associated with a significant decrease in the level of p-catenin whether corrected for the total amount of tissue present (p=0.02) or the amount of epithelium present (p=0.039). Unfortunately transgene expression was not seen in any WT mice treated with the plasmid so it was impossible to consider the effect of APC expression on p-catenin levels in this group.
Figure 38 is a statistical analysis of this same blot when samples were grouped according to treatment arm instead of transgene expression. The elevated base-line level of P- catenin seen when in Min/+ mice treated with 'LIPOFECTAMINE^^ alone' when compared to WT mice treated in the same way (p=0.05) supported the results of section
6.3.1.1.2. Once again there was a trend towards a decrease in the median level of total tissue p-catenin in Min/+ and WT mice treated with APC pCMVp-Neo-Bam, although this failed to reach significance.
6.4 Discussion
6.4.1 Untreated Animals
6.4.1.1 Western Blotting
6.4.1.1.1 APC
Previous work in this group attempted to extract full length and truncated APC from cultured cells using a variety of standard cell lysis buffers, high and low salt lysis buffers and buffers containing urea or the detergent Nonidet-P40. None of these gave good
187 results (Hargest, 1996). The two buffers used here (see Section 6.2.1.1.1 and Appendix 6) were chosen because they had previously yielded full length and truncated APC protein from CRC cell lines (Smith et al., 1993) and full length APC from human colonic mucosa
(Chop et al., 1995). The techniques described for tissue lysis were simple to perform and were carried out under conditions that aimed to minimise proteolysis (see section 6.2.1.1).
However, neither buffer gave adequate Ape extraction when the protein lysates were resolved on SDS-agarose or SDS-PAGE gels. Given that Ape is a large protein and often complexed with other proteins, such as P-catenin, it is possible that the problem lay in inadequate protein extraction, or protein degradation, despite the measures taken to avoid this complication. It is difficult to suggest ways to improve the protein extraction without trying yet another lysis buffer.
Alternatively, or in addition, there may have a problem getting the large Ape protein out of the wells of the denaturing gel and into the resolving matrix. SDS-agarose gels were used in an attempt to circumvent this problem, which had been previously recognised in the context of SDS-PAGE gels (Smith et al., 1993). Unfortunately thin vertical SDS- agarose gels were difficult to make, lacking tensile strength so that the wells collapsed.
Thicker horizontal 3% agarose gels were used instead. It may have been that this created a barrier to protein diffusion, although pore size should be the crucial factor rather than gel thickness. Even running the gels overnight (in a cold-room), m an attempt to increase the movement of large proteins out of the wells, failed to produce a visible band on protein staining or a signal at any weight after immunoblotting with Ab-1.
There is no reason to think that transfer of protein from the agarose gel to the nitrocellulose membrane was a problem. Capillary transfer of APC has been used before
188 (Smith et al., 1993) without incident, and Ponceau S and India Ink staining revealed good transfer of proteins in general following this method of protein transfer.
The final area of difficulty may have been the sensitivity of the primary Ab, Ab-1.
Although successful immunoblotting for APC has been performed using the 'Chop' lysis buffers in combination with Ab-1 (Chop et al., 1995), other papers reporting consistent results with immunoblotting for APC have utilised in-house anti-APC Abs (Smith et al.,
1993, Su et al., 1993a). It is widely felt within the scientific community that commercially available anti-APC Abs are not as reliable as these. However, repeat experiments using in-house ICRF Abs against APC rather than Ab-1 did not alter the results.
Overall it is most likely that several steps in the process, especially tissue lysis and large protein resolution in the gel matrix, were sub-optimal and that Ab-1 sensitivity was not adequate to detect any small quantity of Ape that was present on the nitrocellulose membrane. Following a complete lack of reliable positive results despite adjustment of multiple variables, it was decided to concentrate instead on immunoblotting for p-catenin.
6.4.1.1.2 B-catenin
In comparison with the difficulties encountered in immunoblotting for the Ape protein.
Western blotting for p-catenin proceeded more smoothly. Preparation of the protein samples was straightforward, resolution of protein lysates on SDS-PAGE gels was good and electroblotting gave consistently high quality protein transfer onto the nitrocellulose membrane, with no residual protein in the gels (as judged by staining with Coomassie
Blue after transfer). Initial experiments revealed that a small amount of total protein loading on the gel (5pg) was sufficient to give a strong final signal for p-catenin, P-actin and cytokeratin. The best concentrations of primary and secondary Ab were then chosen 189 in order to give a clear signal with minimal background. The final autoradiographs were fairly consistent, giving clear bands at 10 seconds to 1 minute. The best film was taken as that where the majority of bands were visualised with minimal over or under exposure of any part of the film. This film was used for scanning purposes. In order to check consistency of results, all gels were run in duplicate and each experiment was repeated.
The final p values quoted were from an individual experiment but were only taken as
'correct' if a similar result was seen on repetition.
Stripping and re-probing the membrane gave good results as long as the membrane was not allowed to dry out in between experiments.
In order to keep the number of mice used in initial experiments to a minimum (ten animals) in line with ethical committee requirements, tissue from up to four mice was used to optimise immunoblotting, immunohistochemistry and rtPCR techniques.
Intestinal samples from the remaining mice were used for comparison of base line protein levels. The results obtained were not absolute in terms of nanograms of protein but were relative values, comparing one group to another.
6.4.1.1.2.1 Intestinal Results
Analysis of total cellular p-catenin suggested that there was no statistically significant difference in the level of p-catenin in the recto-sigmoid tissues of Min/+ mice when compared to WT mice, but there was a significant increase in the epithelial level of this protein in the small bowel of Min/+ mice when compared to their WT counterparts.
There may actually be a difference in p-catenin levels in large bowel epithelium that was simply not detected here due to a type II statistical error, having only six recto-sigmoid 190 samples (three Min/+ and three WT). However, in order to allow for the small sample
size, all results were analysed using non-parametric tests and Western blots were carried
out in duplicate with repetition of each experiment on at least one occasion. This technique revealed a significant difference in p-catenin levels in small bowel epithelium,
when only seven mice (four Min/+ and three WT) were considered.
Previous immunohistochemical studies lend some support to these findings. It has been
shown that histologically normal Min/+ small intestinal mucosa exhibits elevated levels
of p-catenin and that this is associated with decreased cellular proliferation and apoptosis
and a decreased rate of enterocyte crypt-villus migration when compared to WT
counterparts (Mahmoud et al., 1997). The same group have demonstrated that treatment
of Min/+ mice with NSAIDS causes a decrease in immunolabelling for p-catenin back
towards the level seen in WT littermates (Mahmoud et al, 1997, Mahmoud et al, 1998b).
However, immunoblotting techniques (often considered less subjective than
immunohistochemistry) were not utilised and the large bowel epithelium was not
considered.
These results are consistent with the hypothesis that mutation in a single Ape allele is
sufficient to produce an elevation in intracellular levels of the oncogene P-catenin within
apparently ‘normal’ Min/+ small bowel but not within equivalent Min/+ recto-sigmoid
tissue. If control of intracellular levels of p-catenin is the important gatekeeper fimction
of Ape, these findings may go some way towards explaining the greater polyp load seen
in Min/+ small bowel. In small bowel epithelium, mutation or loss of the second Ape
allele may not be essential but may offer a further growth advantage. In colonic
epithelium a single Ape mutation does not have such a dramatic effect (possibly due to
local antagonistic factors) and a second Ape mutation may be absolutely required to
191 produce the change in p-catenin that is needed for progression of the adenoma-carcmoma sequence.
6.4.1.2 Immunohistochemistry
The Ab used in immuno labelling Ape was directed towards its carboxy-terminal. This meant that a positive signal was generated in both Min/+ and WT mice. The initial assumption, that there would be a quantitative difference in signal between the two groups was not home out, as any differential result was lost in the amplification steps.
Therefore this was not a good basis for the assessment of successful gene therapy.
Immunohistochemistry for p-catenin appeared to confirm the results of section 6.2.1.1.
However, the quality of the sections was poor so it was decided to concentrate on the best all round method, immunoblotting for p-catenin, as a way to measure any effect of APC gene therapy at the protein level.
6.4.2 Treated Animals
The tiny amounts of intestinal tissue obtained from four week old mice (especially recto sigmoid samples) meant that there was only enough to provide two equal samples, one for rtPCR and one for protein analysis. Although a small number of the designated ’protein' samples were required for repeat rtPCR experiments, none of these came from tissues expressing the transgene.
In clinical trials it is conventional to group by 'intention to treat' or treatment arm.
However, given the small number of mice treated here it was felt that grouping animals according to the presence or absence of transgene expression in the other half of that tissue sample allowed a clearer assessment of the downstream effects of successful APC
192 gene transfer on intestinal levels of P-catenin. This was especially the case in the analysis
of small bowel samples following gavage where the problems of gene transfer were more
obvious (see chapter 4). Grouping by transgene expression gave an idea of the results that might be achievable once gene transfer is optimised.
Given that immunohistochemistry here and elsewhere has demonstrated that Ape and p-
catenin are almost exclusively epithelial proteins, correction of the P-catenin signal for
the amount of total tissue present (correcting for p-actin) was expected to give similar results to correction for the amount of epithelial tissue present (correcting for
cytokeratin). This was observed. The fact that some experiments reached statistical
significance when corrected in one way but not in the other suggests that the true size of
the change in P-catenin lies somewhere between these results. Sample numbers may have
obliterated this small discrepancy.
6.4.2.1 Rectal Trial
Although it appears that there is not a significant increase in pre-neoplastic recto-sigmoid
levels of p-catenin in the Min/+ mouse (as discussed in section 6.3.1) treatment with APC
pCMVP-Neo-Bam and transgene expression did appear to be associated with a reduction
in the level of p-catenin in both genotypes (Min/+ and WT). This effect reached
significance in Min/+ mice when the mice were grouped according to transgene
expression (p=0.019) and only just failed to be significant when a treated mouse not
expressing the transgene was included in the positive treatment group (p=0.053). If
treatment failure were due to loss of the enema before it could be absorbed, such a
procedural failure would be obvious in treating human disease and could be addressed by
repeated dosing or administration of therapy after a spontaneous evacuation.
193 Therefore it appears thatAPC gene enemas may be a useful way of accessing the rectum in patients with FAP and may provide a means to lower abnormally raised P-catenin levels. The effect this has on polyp formation in the rectal stump remains to be seen. The
Min/+ mouse rectum is clearly not a good model in which to study this, but the Min/+ small bowel appears to be a better disease model
It was interesting that there also appeared to be a small but non-significant effect of APC gene transfer, or treatment with APC pCMVp-Neo-Bam, on P-catenin levels in WT mice
(p=0.386 when corrected for P-actin, p=0.564 when corrected for cytokeratin). If this were a true effect, APC gene enemas could be a useful treatment for non-FAP patients at increased risk of CRC if p-catenin abnormality was a recognised feature of disease. Once again larger studies are needed to clarify these results.
6.4.2.2 Gavage Trial
The effect of transgene expression on p-catenin level in the small bowel epithelium of
Min/+ mice was more pronounced (p=0.02-0.039). Unfortunately this effect was lost when analysed for treatment arm, suggesting that APC gene therapy will only be a feasible treatment for the duodenal manifestations of FAP when a better rate of gene transfer can be guaranteed. In the future, this may be achieved by coating the APC-
liposome complex so that it passes through the stomach undamaged (Niedzinski et al.,
2000), but the effect of small bowel contents on liposomes is not yet clear. This is currently being considered within the group. Alternatively the problem may be addressed with the introduction of alternative gene delivery vectors. Attenuated salmonella strains have been used as vectors for oral gene therapy against murine B-cell lymphoma, as they are not destroyed in the digestive tract and deliver the gene of interest to lymphoid tissue within the gut (Urashima et al., 2000). AAV vectors have also been used successfully in
194 upper GI gene therapy (During et al, 1998). Either of these may provide improved delivery of^PC gene therapy.
Overall both the rectal and gavage trials revealed interesting results, showing that APC gene therapy induced a reduction in intestinal epithelial levels of P-catenin in Min/+ mice as long as the environmental conditions for good gene transfer were present. This effect may have also been present to a lesser degree in WT mice, whose base-line protein levels were normal.
195 Figure 25-Estimation of Protein Concentration
Known Standard Absorbance Concentration at 562nm pg/ml 0 0 200 0.124 400 0.236 600 0.345 800 0.414 1000 0.53
Linear Regression of Spectrophotometric Analysis
C 0.6 ^ 0-5 « 0.4 S 0.3 S 0.2 .o
E 0 < 0 200 400 600 800 1000 1200 Protein concentration pg/ml
Sample Name Absorbance Sample at 562nm Concentration pg/ml 0.287 523.3374767 0.277 504.155989 0.264 479.220055 0.238 429.3481871 0.241 435.1026334 0.293 534.8463693 0.275 500.3196915 0.176 310.4229635
96 Figure 26- Immunblots of Recto-Sigmoid Tissue Samples Taken From
Untreated Mice a) Probed For p-catenin
220kD —
97.4kD —
M M M W W W
b) Probed For p-actin
66kD
46kD M M M W W W
c) Probed For Cytokeratin Eight
66kD
46kD - M M M W W W
M = Min/+ sample
W = WT sample
197 Figure 27- Statistical Comparison of P-catenin Levels in The Recto-
Sigmoid of Untreated Min/+ and WT Mice
1.6
P-catenin 1.4 signal 1.2 adjusted for 1.0 P=0.827 p-actin 0.8 m 0.6 signal 0.4
N=
1.6 p-catenin 1.4 signal 1.2 adjusted for 1.0 cytokeratin P=0.827 0.8 !>sr eight signal 0.6 0.4
N=
□ Min/+ Mice WT mice
— Median value of P-catenin
The height of the box is the interquartile range, which contains 50% of the values.
The whiskers extend to the highest and lowest values. There were no outliers.
N= number of mice used for data collection
198 Figure 28- Immunblots of Proximal Small Bowel Tissue Samples Taken
From Untreated Mice a) Probed For p-catenin
M M W W W M M b) Probed For p-actin
36.5kD — M M W W W M M c) Probed For Cytokeratin Eight
58kD -
45kD - M M W W W M M
M = Min/+ sample
W = WT sample
199 Figure 29- Statistical Comparison of Total and Sub-cellular Levels of P-
catenin in The Proximal Small Intestine of Min/+ and WT Mice
2.4 2.2 p-catenin 2.0 signal 1.8 adjusted for 1.6 P-0.289 1.4 P-act in 1.2 signal 1.0 0.8 0.6
N=
P-catenin 3.5 3.0 signal 2.5 adjusted for 2.0 P-0.034 cyto keratin 1.5 eight signal 1.0 0.5
N=
□ Min/+ Mice WT mice
Median value of p-catenin
The height of the box is the interquartile range, which contains 50% of the values.
The whiskers extend to the highest and lowest values. There were no outliers.
N= number of mice used for data collection. 200 Figure 30-Haematoxylin and Eosin Staining of Min/+ Intestinal Frozen Sections
a) Small Bowel x20 magnification
4
ȣ b) Recto sigmoid x20 magnification
■ifê
:'C. ^ S u ‘;0 -> '• •*' :
201 Figure 31-Immunoblotting APC In Intestinal Frozen Sections
a) Min/+ Recto sigmoid x20 b) WT Recto sigmoid x20
c) M in/+ Small Bowel x20 d) WT Small Bowel x20
: # mge,..:.
202 Figure 32-Immunoblotting P-Catenin in Intestinal Frozen Sections
a) Min/+ Recto sigmoid x20 b) WT Recto sigmoid x20
»
c) Min/+ Small Bowel x20 d)WT Small Bowel x20
203 Figure 33-Immunblots of Recto-Sigmoid Tissue Samples Taken From
Rectaiiy treated Mice a) Probed For p-catenin
116kD -
84kD - - - ++ + ++ + -- - - + + + +
b) Probed For P-actin
45kD
+ + + + + + - - - - +4-4-4-
c) Probed For Cytokeratin Eight
45kD - - - + ++ + ++ -- - - + + + +
- = Min/+ sample with no transgene expression in the other half of the recto-sigmoid
"F = Min/+ sample expressing the transgene in the other half of the recto-sigmoid
- = WT sample with no transgene expression in the other half of the recto-sigmoid
4" = WT sample expressing the transgene in the other half of the recto-sigmoid
204 Figure 34- Alterations in Recto-Sigmoid Levels of p-catenin In Response
To Transgene Expression
1.3 n P = 0 .3 8 6
P-catenin 1.2 P = 0.67 signal adjusted for 1.1 - P-actin
signal 1.0
0.9
N=4
30 -r p-catenin
signal 20 - P=(),083 P=().019 adjusted for
- i cytokeratin 10 ■ eight signal
N=4
^ Min/+ sample with no transgene expression in the other half of the recto-sigmoid
I I Min/+ sample expressing the transgene in the other half of the recto-sigmoid
9 WT sample with no transgene expression in the other half of the recto-sigmoid
I I WT sample expressing the transgene in the other half of the recto-sigmoid
Outlier, specimen number 6.
N= number of mice used for data collection
P values above boxes are in comparison with Min/+ samples not expressing the transgene
205 Figure 35- Alterations in Recto-Sigmoid Levels of p-catenin In Response
TM To Treatment with APC pCMVP-Neo-Bam or LIPOFECTAMINE
1.3 n P = 0.724
(3-catenin 1.2 signal P = 0 .2 I0 adjusted for 1.1 P-actin signal 1.0
0.9
N=3
30 -1 P-catenin signal 20 - P = 0.077 adjusted for P = 0.053 cytokeratin 10 - eight signal
N=3 7
Sample from a Min/+ mouse treated with 'LIPOFECTAMINE'^ alone'
Sample from a Min/+ mouse treated with APC pCMVp-Neo-Bam
Sample from a WT mouse treated with 'LIPOFECTAMINEalone'
Sample from a WT mouse treated with APC pCMVp-Neo-Bam
N= number of mice used for data collection
P values above boxes are in comparison with Min/+ mice treated with
'LIPOFECTAMINE^'^ alone'
206 Figure 36-Immunblots of Tissue Samples Taken From The Proximal
Small Bowel of Mice Treated By Gavage a) Probed For p-catenin
116kD -
84kD - ...... 4- + + ------b) Probed For P-actin
45kD —
------4- 4- 4- c) Probed For Cytokeratin Eight
58kD -
45kD -
+ 4-4-
- = Min/+ sample with no transgene expression in the other half of the tissue sample
+ = Min/+ sample expressing the transgene in the other half of the tissue sample
- = WT sample with no transgene expression in the other half of the tissue sample
+ = WT sample expressing the transgene in the other half of the tissue sample
207 Figure 37- Alterations in Proximal Small Bowel Levels of p-catenin In
Response To Transgene Expression
1.6 “
1.4 - P-catenin 1.2 - signal P = 0.004 1.0 - adjusted for 0.8 - P=0.02 p-actin 0.6 - % signal 0.4 - 0.2 - 0.0 -
N= 6
P-catcnin 5 - signal 4 - adjusted for 3 -
P = 0.039 cytokeratin 2 - P = 0.004 eight signal 1 -
N = 6
^ Min/+ sample with no transgene expression in the other half of the sample
Outlier, specimen number 1.
I I Min/+ sample expressing the transgene in the other half of the sample
^ WT sample with no transgene expression in the other half of the sample
No WT mice expressed the transgene in the proximal small bowel
N= number of mice used for data collection
P values above boxes are in comparison with Min/+ samples not expressing the transgene
208 Figure 38- Alterations in Levels of p-catenin In The Proximal Small
Bowel In Response To Treatment with APC pCMVp-Neo-Bam or
TM LIPOFECTAMINE
1.6 1.4 P=0.071 1.2 P-catenin 1.0 P =0.05 signal 0.8 adjusted for 0.6 P-actin 0.4 signal 0.2 0 N=3
P-catenin
signal P = 0.606 adjusted for cytokeratin P = 0.05 eight signal
N=3 6 3 3
Sample from a Min/+ mouse treated with 'LIPOFECTAMINE'^ alone'
I I Sample from a Min/+ mouse treated with APC pCMVp-Neo-Bam
-F" Outlier, specimen number 5
HI Sample from a WT mouse treated with 'LIPOFECTAMINE'^ alone'
B Sample from a WT mouse treated with APC pCMVp-Neo-Bam
N= number of mice used for data collection
P values above boxes are in comparison with Min/+ mice treated with
'LIPOFECTAMINEalone'
209 Chapter 7
Discussion
210 7.1 Rationale Underlying Study Design
Loss or mutation of a tumour suppressor gene or mutation of a proto-oncogene initiates neoplasia. Corrective gene therapy aims to restore equilibrium in 'abnormal' cells with the administration of additional 'normal' copies of an affected tumour suppressor gene or by blockade of an affected oncogene (section 1.9.3.3). The majority of research protocols to date have considered corrective gene therapy with WT p53, as this well characterised tumour suppressor gene is known to be mutated in a wide range of tumours.
Proof of principle studies with WT p53 gene therapy have clearly demonstrated the benefit of corrective gene therapy in cellular systems, animal models of lung cancer and in human lung disease (sections 1.9.3.3.1 and 1.9.3.3.3). Given the prevalence o f APC mutations in sporadic CRC (section 1.4) and its position as an initiator of the adenoma- carcinoma sequence (section 1.2), it seems an ideal target for corrective gene therapy for
CRC. Those individuals affected by FA? are a sub-group particularly likely to benefit as their disease is always caused by a germline mutation in this same gene. FA? is also a good disease in which to evaluate cancer gene therapy as those regions of bowel most at risk of malignant transformation (the rectal stump after IRA and the duodenum) are the most easily accessible for local treatment.
7.1.1 Liposome Mediated Gene Therapy
To date, viral mediated gene therapy has been the main focus of attention due to its ability to provide good transfection efficacy. However, in the light of a recent death following an adverse immune reaction to an adenoviral vector (Lehrman, 1999), and concerns surrounding the possibility of recombination events in all viral vectors, interest in ‘safe’ non-viral methods of gene delivery has been rekindled. Liposomes are the most widely used non-viral gene vectors (section 1.9.2.2.1), and they have already proved useful in
211 clinical trials of cancer gene therapy (Nabel et al, 1993) and cystic fibrosis gene therapy
(Caplen et al., 1995a).
The large DNA carrying capacity of liposomes is also an advantage when considering the transport of a large gene, such as APC.
7.2 Summary of Experimental Findings
As indicated in Section 1.9.3.3.4.2, this study set out with four main objectives
• To investigate the levels of APC transgene expression that could be obtained in the
rectum and proximal small bowel of Min/+ mice following local administration of a
liposome/ DNA complex that has previously been shown to yield high level
expression in the rectum of WT mice.
• To determine the phenotypic effects resulting from such transgene expression, at a
clinical and molecular level.
• To investigate the range and level of tissue expression resulting from intra-peritoneal
delivery of the same complex in a group of WT mice.
• To determine the safety of APC pCMVp-Neo-Bam at all times.
7.2.1 Rectal Administration of APC pCMVp-Neo-Bam
High level recto-sigmoid expression of the transgene was seen following a single enema treatment (section 4.2.1.1). This may have been due to the relatively large volume of enema (0.5ml) distending the rectum, increasing the area of surface contact and causing local uptake of the transgene within the cells that came into contact with the lipid/DNA complex (Sandberg et al., 1994). Transgene expression was not seen more proximally, in the colonic mucosa that did not come into direct contact with the vector (section 4.2.1.2).
212 There was no systemic expression o ïA P C (section 4.2.1.3), a useful safety feature. Only one mouse within the rectal trial became ill and this did not appear to be as a result of the treatment (section 5.2.1).
Given the low recto-sigmoid polyp load of Min/+ mice, the young age of these mice at treatment, the use of only a single treatment dose and the short time period between treatment and death, the absence of a change in polyp load following treatment and transgene expression (section 5.2.1) was not surprising.
Turning to the molecular effects of treatment with APC pCMVp-Neo-Bam and/or transgene expression, it was interesting to see that untreated Min/+ mice did not have a statistically significant elevation in pre-neoplastic epithelial levels of p-catenin in the recto-sigmoid when compared to their WT counterparts (sections 6.3.1.1 and 6.3.1.2), but transgene expression and treatment with APC still appeared to decrease the median level of epithelial p-catenin (p=0.019 and p=0.053 respectively). There appeared to be a similar effect after the treatment of WT mice, who were presumed to have 'normal' levels of p- catenin.
7.2.2 Administration of APC pCMVp-Neo-Bam By Gavage
Only a third of the mice treated with APC pCMVp-Neo-Bam expressed the transgene in the stomach, and only a quarter continued to express it in the proximal small bowel
(section 4.2.2.1). This was a poor level of expression when compared to that achieved in the rectum. This difference may have been due to incomplete distension of the stomach as a result of a decrease in treatment volume (section 3.2.9). Alternatively, the effect of the hostile physical environment on the lipid/DNA complex (section 4.3.5.1) or an intrinsic problem with uptake into small bowel epithelium may have been important. The effect of 213 different pH and bile salt concentrations on transfection efficiency of APC pCMVp-Neo-
Bam into small bowel cell lines is currently being considered within the group in an attempt to address this problem.
Once again there was no systemic expression of APC (section 4.2.2.2), a useful safety feature. None of the mice involved in the gavage trial suffered any ill effects.
It was not surprising that no significant clinical change occurred in the proximal small intestine as a result of either treatment with APC pCMVp-Neo-Bam or transgene expression. Although Min/+ mice develop the majority of their polyps in the small intestine, these mice were still subject to the other limitations seen in their littermates in the rectal trial, discussed above.
Immunoblotting for P-catenin in proximal small bowel tissue lysates from untreated
Min/+ and WT mice revealed a statistically significant elevation in pre-neoplastic epithelial levels of p-catenin in the Min/+ mice when compared to their WT counterparts
(p=0.034, section 7.3.1.1.2). Transgene expression in the proximal small bowel of Min/+ mice correlated with a decrease in the median level of epithelial p-catenin (p=0.002 to p=0.039, section 7.3.2.2), bringing it down into the range seen in WT mice.
Thus overall, it seems that transgene expression is associated with, and probably causes, a decrease in cellular levels of p-catenin. Given that abnormal p-catenin regulation is felt to underpin the neoplastic effect of APC mutations in FAP and the majority of sporadic adenomatous polyps, this finding supports the concept of local corrective gene therapy as a treatment for intestinal disease in patients with FAP.
214 Future studies may concentrate on the development of more efficient vectors, especially
in the upper intestinal tract (see section 6.4.2.2), repetition of these results in larger
cohorts of animals and/or small human trials and an analysis of the effect of repeated
administration of the gene.
7.2.3 Intra-peritoneal Administration of APC pCMVp-Neo-Bam
It is known that the majority of FAP associated intra-abdominal desmoids arise from
fibroblastic cells in the mesentery (section 1.2.2.3), but the relationship between peritoneal fibroblasts and desmoids is unclear. FAP desmoids are true clonal neoplasms with deletion or mutation of the WT APC allele accompanying the germline mutation
(Dangel et al, 1994, Palmirotta et al, 1995, Miyaki et al, 1993). Assuming that APC
exerts its effect in connective tissue in a similar manner to that described in the intestine,
corrective APC gene therapy instilled into the abdomen of FAP patients at the time of
surgery, or regular administration of intra-peritoneal gene therapy in those individuals at high risk of desmoids (section 1.3.1), might help to decrease the incidence of this disease.
Therefore, studies are underway to consider the effect of APC gene transfer into
fibroblasts, keloids (immortal fibroblasts) and desmoid cells in culture, paying particular
attention to the effect this has on cell growth and p-catenin expression.
This study demonstrated that intra-peritoneal delivery of APC pCMVP-Neo-Bam leads to
a high level of transgene expression in the mesentery and peritoneum (section 4.2.3.1),
frequent expression in the liver (section 4.2.3.3) and all segments of intestine (section
4.2.3.2), without any ill effects in the mice. This is an important starting point for in vivo
studies of APC gene therapy for desmoid disease. The next avenue of investigation, in
line with the in vitro experiments outlined above, would be to consider the clinical and
molecular effects of such gene transfer in a murine model of desmoid disease.
215 7.3 Limitations of The Study
Although the Min/+ mouse is the most widely used, and probably the best characterised,
animal model of FAP (section 1.8.1) its relative lack of rectal polyps meant that it was not
an ideal subject in which to study the clinical effectiveness of 'gene enemas'. Considering
another treatment end-point circumvented this problem. In the future, Apc^^' Msh2'^' mice
might be a better model in which to study rectal polyposis, as they develop many APC
deficient rectal polyps in a short period of time (section 1.8.3).
In order to obtain ethical committee approval for the study, it was necessary to show that
the minimum number of mice needed to give useful results was being used in each
treatment arm. Ten mice were used in setting up the protocols and in base-line
comparison of protein levels. Eight Min/+ mice and four WT mice were treated with APC
pCMVp-Neo-Bam (with four Min/+ and four WT controls) in each of the intestinal
therapy trials. Eight WT mice and four WT controls were used in the intra-peritoneal
study. In retrospect, larger group sizes may have clarified the results in those areas where
an alteration in p-catenin levels was suggested but did not reach significance.
7.4 The Future of Corrective Gene Therapy
The concept of cancer gene therapy was initially greeted as a panacea to eliminate all
conventional therapies. However, it has become increasingly clear that sub-optimal
efficiency and accuracy of gene delivery are major stumbling blocks that have still to be
overcome, and that there are important safety issues to be considered. Effective corrective
gene therapy is likely to require a specific combination of genes and genetic techniques
for each tumour type, and may never be as widely used as other forms of gene therapy
(sections 1.9.3.1-1.9.3.2). Having said this, great strides are being made with the
216 development of more effective and safer vectors and in the identification of new targets for genetic manipulation.
7.4.1 Viral Vectors
Packaging cell lines that split the packaging genes of retroviral vectors between two plasmids (Markowitz et al, 1988) have been introduced in an attempt to reduce the number of recombination events to a minimum.
'Gutless' or high capacity adenoviral vectors, have almost all viral genes deleted except the packaging sequence and the terminal repeats. This means that the insert capacity of these vectors is increased (up to 30Kb) with a simultaneous decrease in vector immunogenicity (Schiedner et al., 1998, Thomas et al., 2000).
Oral administration of adenoviral protein extracts, or whole inactivated virus, also reduces the development of anti-adenoviral neutralising Ab and down-regulates the
cytotoxic T-cell response to the vector (Ilan et al., 1997, Kagami et al, 1998). Concurrent treatment with immunosuppressive chemotherapeutic agents has a similar effect (Bouvet
et al, 1998b). These approaches may prove useful in repeated administration of adenoviral vectors.
Some viruses have an intrinsic oncolytic effect. Gene attenuated replication competent
adenoviruses are now available which selectively replicate in and destroy cancer cells
(Heise et al, 1997). The effectiveness of this strategy is being evaluated in clinical trials.
The introduction of exogenous therapeutic genes into these viruses offers yet another layer of anti-tumoural effect (Russell, 1994). For example, replication competent adenovirus carrying HSV tk was given to mice bearing CRC xenografts. Half of the mice
217 were subsequently treated with ganciclovir, and it was seen that the addition of this
'suicide gene therapy' significantly enhanced the antitumour effect of the virus alone
(Wildner et al, 1999). The obvious concern with these systems is the maintenance of a safe vector, which will not damage 'normal' tissue.
Lentiviruses, such as the human immunodeficiency virus type 1 (HIV-1) utilise the nuclear import mechanisms of non-mitotic cells to cross the nuclear membrane. Thus they can integrate into the DNA of quiescent cells (Naldini et al., 1996). This makes them a potentially useful vector when persistent gene expression is required (Trono, 2000). Work is on-going looking at ways of modifying the HIV-1 virus to make it a safe and efficient vector for gene transfer (Trono, 2000).
7.4.2 Non-Viral Vectors
Physical approaches, such as particle bombardment (the 'gene gun') (Yang et al., 1990) and electroporation (Rizzuto et al., 1999), allow naked plasmid DNA to bypass the
'normal' cellular entry mechanisms for macro molecules. This reduces lysosomal degradation and thus enhances transgene expression.
Alternatively, protective interactive and non-condensing (PINC) polymers, such as polyvinylpyrrolidone (PVP) have been complexed with plasmids and introduced into muscle or subcutaneous tissue. These polymers protect the plasmid against enzymatic
degradation and facilitate slow release of the DNA within the target tissue (Anwer et al,
1999).
Although liposomes offer a potentially safe, plentiful and cheap gene delivery mechanism, they still do not come close to the efficiency of gene transfer offered by viral
218 vectors. Therefore, time and effort is being expended in several fields of basic research in an attempt to improve transfection efficiency. The in v/vo performance of lipofection depends partly on the route of administration. For example, serum induces lipid vector disintegration, leading to DNA release and degradation (Li et al., 1999). This effect is especially prominent in the presence of the neutral fusogenic lipid dioleoylphosphatidylethanolamine (DOPE), a constituent o f LIPOFECTAMINE™. In contrast DOPE enhances lipofection across an epithelial surface such as the trachea.
Therefore, the production of individual lipid vectors with components tailored to the route of administration may become a reality in the future.
Additional areas of interest include techniques to improve the targeting of liopsomes to tumour cells (discussed below), the incorporation of fiisogenic peptides into the lipid complex to enhance cytoplasmic release of the DNA, and ways to improve nuclear uptake
of the plasmid DNA from the cytosol.
Cationic polymers, such as polyethylenimines, have recently been used as gene delivery
vectors (Boussif et al., 1995). They produce efficient condensation of plasmid DNA and prevent endosome-mediated destruction of the complex
7.4.3 Hybrid Vectors
Viral manipulation and recombinant DNA technology now allows the combination of
useful functions from different vectors in one hybrid.
Viral hybrids have been produced where an adenoviral vector (providing good
transduction capabilities) delivers an Epstein Barr virus episome (that can replicate intra- cellularly independent of mitosis) to infected cells (Tan et al., 1999). Adeno-retroviral
219 chimeras are also in development. Here, one or more adenoviral vectors carry adenoviral packaging genes and a transcomplementary retroviral protein of interest in to a cell. A useful retroviral vector is then produced and packaged in situ. This can in turn transduce target tissues (Caplen et al., 1999).
Non-viral hybrids have been generated with the addition of cationic polymers, such as poly(L-lysine) or protamine, to liposomal vectors. This enhances transfection efficiency, through a reduction in size of the complex and increased resistance to nuclease activity.
(Gao and Huang, 1996). However, the transfection ability of these vectors is still limited by endosome/lysosome destruction
Molecular conjugates, like transferrin-polylysine/DNA complexes, have also been linked to the scaffold of the adenovirus capsid (Wagner et al., 1992). Adenoviral cell entry mechanisms provide the conjugate with an endosome escape route, so ensuring improved transfection efficiency. At the same time the position of the DNA external to the virus removes size constraints. In addition, since the gene of interest is bound to the polymer in a sequence independent mechanism, DNA of any design can be utilised.
Ultimately the aim of this approach is to combine the safety features of non-viral vectors with the transfection efficiency of viral vectors.
7.4.4 Targeting Vectors
Systemic treatment regimes will require gene therapy vectors to be protected whilst in transit to the target tissue and to be able to selectively target and enter the tissue of interest.
220 Adenoviral vectors, probably the most widely used vector in human clinical trials, normally enter a wide variety of cells through an interaction between the knob domain of the adenoviral fibre protein and the Coxsackie adenoviral receptor, CAR, on the cell surface. Re-targeting the virus can be achieved by blocking the fibre knob with an Ab, the
Fc portion of which is attached to another ligand that is specific for tumour cells (Haisma et al., 1999). Alternatively, the knob can be removed and replaced with another protein.
In a slightly different vein, Niedzinski et al have attempted to address this problem with respect to the intestinal delivery of liposome mediated gene therapy. They have produced liposomes with superficial modified cholic-acid esters that simultaneously protect the
DNA fi"om digestion by deoxyribonucleases and enhance entero-hepatic gene delivery
(Niedzinski et al., 2000).
Transcriptional targeting of vectors, increasing the specificity of gene therapy, can be achieved with the use of tissue or tumour specific promoters. For example VDEPT has been used in c-erhB2 positive breast cancer cells when the prodrug 5-FC was placed under the c-erhB2 promoter (Harris et al., 1994). This enhanced tumour specific cell kill.
Adenoviral vectors carrying HSV tk under a CEA promoter have also been compared to the same construct under a CMV promoter in animal models of primary and metastatic
CRC. The CEA promoter increased the specificity of tumour cell killing although its efficacy was not as great as that seen with the CMV promoter (Konishi et al., 1999, Brand et al., 1998). However, such transcriptional targeting may not ultimately prove to be as usefiil as surface targeting which minimises difference between the dose given to the patient and the dose taken up by the tumour.
221 7.4.5 New Vectors
Certain bacteria, such as salmonella and E.Coli, have adapted to survive and flourish
within the intestine. Therefore, these may prove to be good vectors for intestinal gene
therapy. Attenuated salmonella strains have already been used as vectors for oral gene
therapy against murine B-cell lymphoma (Urashima et al., 2000) and are being considered
in the treatment of other diseases .
Individual viral proteins have also been considered as potential vectors. VP22 is a HSV
type 1 transport protein (Elliott and Meredith, 1992) capable of highly efficient, gap junction independent, movement from one virally infected, or VP22 transfected, cell to
approximately 200 neighbouring cells (Elliott and O'Hare, 1997). The gene encoding
VP22 has been linked in frame to HSV tk, and transfected into Cos cells. These cells
produce a chimeric protein that is able to move into adjacent cells. Following treatment of
the cells with ganciclovir, the bystander effect with the VP22-HSV tk fusion gene is
much greater than with HSV tk alone; (Elliott and O'Hare, 1999, Dilber et al., 1999). In
order to assess the clinical benefits of using purified VP22 as a gene delivery system,
consideration is being given to the physiological functions of this protein and the clinical
effect of its accumulation in body tissues.
The perfect gene vector would not only deliver one or more genes of interest to a target
tissue at high concentration, but it would orchestrate gene expression in a physiological
manner. In theory, a mammalian artificial chromosome (MAC) or human artificial
chromosome (HAG) would be able to carry several genes and their controlling sequences,
it would be maintained within the nucleus of the target cell in parallel to the intrinsic
nuclear material and replicate concurrently with the target cell.
222 The Yeast artificial chromosome (YAC) is already a widely used tool in molecular biology. A linear mitotically stable YAC requires a minimum of three chromosomal elements, a centromere (that binds to microtubules during mitosis), telomeres (that control cell cycle number and cell ageing) and an origin of replication (Beach et al., 1980,
Bloom and Carbon, 1982, Shampay et al., 1984). However, there have been difficulties transferring this technology from prokaryotic to eukaryotic systems. This is partly because of the size of human chromosomes (50-250 megabases, approximately 100 times larger than their yeast counterparts), and partly due to difficulties in manufacturing a suitable centromeric structure.
However, the first generation HACs were finally produced in 1996. They were linear microchromosomes (up to ten megabases) containing a synthetic centromere, telomeric
DNA and a small amount of genomic DNA. They were maintained stably in culture for six months (Harrington et al, 1997). Further development is on going.
7.4.6 New Forms of Gene Therapy
7.4.6.1 Antiangiogenic gene therapy
The growth of all tumours is dependant upon an adequate blood supply (Folkman, 1990).
Therefore, inhibition of expression of endogenous angiogenic factors such as vascular endothelial growth factor (VEGF), or overexpression of antiangiogenic molecules, such as angio statin, could provide an effective therapy for a wide range of tumours. Gene therapy is potentially an ideal way to deliver such treatment as it offers the prolonged expression of the therapeutic proteins necessary to maintain inhibition of tumour growth.
Targeted gene therapy may also minimise the effect on systemic vasculature.
223 Once again, this therapy will not fulfil its promise until vector technology and targeting techniques are optimised. It may also be that tumour sensitivity to individual angiogenic factors changes with time (Bergers et al., 1999) and a combination of anti-angiogenic agents may prove to be the best way forward in this field.
T.4.6.2 Targeted Gene Repair
With an increasing awareness of the importance of mismatch repair mechanisms in the maintenance of a 'normal' cell phenotype, it has been suggested that these same mechanisms could be put to use to repair the simple genetic mutations that underlie many cancers. The main advantage of gene repair is that it offers to correct the gene in situ, so that it remains under its normal regulatory control. Many techniques are under investigation, chimeraplasty and ribozymes, two promising approaches, are considered here.
Artificial RNA-DNA chimeras are produced when a DNA region with a single ‘correct’ nucleotide is placed between two RNA bridges complementary to the target gene. This chimera is then transfected into the target cell, containing the mutant gene, and binds to it.
MMR enzymes recognise the single base pair mismatch between the endogenous mutant gene and its DNA copy within the chimera. Because the RNA within the chimera is modified to protect against enzymatic degradation, the enzymes remove and replace the endogenous mutant DNA using the sequence information provided by the chimera as the
'correct' template (Gura, 1999). Chimeraplasty has been used to correct a single-base mutation causing sickle-cell anaemia, in cells from patients with this disease. Ten percent of targeted cells expressed the 'normal' gene (Cole-Strauss et al., 1996). However, before it can be promoted in human trials, more work is needed to identify factors controlling the repair process and to ensure that other regions of the genome will not be subject to
224 unnecessary, possibly deleterious, alteration. The current efficiency of gene repair also needs to be addressed.
Ribozymes are enzymes that target mutant RNA at pre-determined sites, so preventing its translation. The may also replace a mutant stretch of RNA with a correct RNA sequence
(Sullenger and Cech, 1994). This technology is at a very early stage, but one obvious problem is the need for repeated treatments in therapies that target RNA rather than DNA.
7.5 The Future Role of^PC Gene Therapy In Colorectal Cancer
Much work has been carried out in an attempt to understand the genetic basis of CRC. By the late 1980s, the importance of APC mutations in the classical adenoma-carcinoma pathway had been recognised. During the 1990s, the cellular role of APC, (section 1.6) was elucidated and it became clear that some of the CRCs that do not follow the classical pathway, develop due to defects in mismatch repair (section 1.1). As the various pathways are revealed, new targets for genetic manipulation continue to arise. Recent
work suggests that TCF-4, one of the important down stream effectors of the APC/p- catenin pathway is also a target gene for MMR mutations in HNPCC (Duval et al., 1999).
In other words, alteration in transcriptional regulation, particularly via TCF-4, is a common pathway to neoplasia, regardless of the initial genetic defect. It is likely that these transcriptional regulators will soon become targets for gene therapy and/or gene repair.
APC gene therapy may yet prove useful in TAP, in the prevention or reduction of rectal,
duodenal and possibly desmoid disease. It may even be a part of the armamentarium against other CRC syndromes or sporadic disease in high-risk individuals. However, the
incomplete transfection currently achievable makes it most likely that it will be used
225 either as an adjuvant treatment to enhance the effect of other treatment modalities (such as surgery, NSAIDS, chemotherapy and/or radiotherapy) or as a chemopreventive agent in early stage, low bulk disease.
226 Chapter 8
Appendices
227 Appendix 1-Mouse Genotyping
Ear Lysis Buffer
28mM Sodium chloride (NaCl)
55mM Tris (hydroxymethyl) methyl ammonium chloride, pH 8
0.1 % Sodium dodecyl sulphate (SDS)
PCR Reaction Conditions
Step 1 94°C 5 minutes xl Initial dénaturation
Step 2 94°C 1 minute Dénaturation
58°C 2 minutes > - x30 Annealing
72°C 3 minutes Extension
Step 3 72*C 10 minutes xl Final extension
228 Appendix 2- Agarose Gel Electrophoresis
50 X TAE (Tris acetate EDTA) Buffer
242g Tris (hydroxymethyl) aminomethane base
57.1ml Glacial acetic acid
100ml 0.5M EDTA (ethylene diamine tetraacetic acid) pH 8
Made up to 11 with ddHiO. Autoclaved and stored at room temperature. Diluted with H 2 O for use at IxTAE.
Gel loading Buffer (Orange G)
27.5ml Glycerol
22.5ml Ix Tris EDTA Stored at room temperature.
0,1 g Orange-G powder
DNA ’Ladders’
Ikb Ladder (Sigma-Aldrich Company Ltd, Poole, Dorset, UK)
0.5-3kb in steps of 0.5kb, up to 6kb in steps of Ikb and up to lOkb in steps of 2kb. For easy reference on an agarose gel the Ikb band is brighter than the other bands.
lOObp Ladder (Promega Ltd) lOO-lOOObp in steps of lOObp, with an additional band at 1500bp The 500bp band is brightest.
229 Appendix 3-Plasmid Preparation and Purification
Solutions For Culture of Bacteria Containing Plasmids
Luria-Bertani (LB) Broth
20g commercially available LB base (Sigma-Aldrich Company Ltd, Poole, Dorset, UK) was made up to 11 in ddH20 and autoclaved. It was stored at 4®C.
LB Plates
LB broth was made and, after autoclaving, 15g agar per litre was added. The solution was cooled to 50®C in a waterbath. At this point ampicillin was added to a final concentration of lOOpg/ml. Approximately 20 ml of solution was poured onto each sterile petri dish and allowed to cool.
lOx Tris EDTA (TE)
10ml IM Tris, pH 8
2ml 0.5M EDTA, pH 8
Made up to 11 with ddHiO, and stored at room temperature. Diluted for end use to IxTE.
Qiagen-Maxi Kit
Buffer PI (Resuspension Buffer)
50mM Tris (hydroxymethyl) methyl ammonium chloride, pH 8.0 lOmM EDTA lOOpg/ml RNAse A Stored at 4°C
230 Buffer P2 (Lysis Buffer)
200inM Sodium hydroxide (NaOH)
1% SDS Stored at room temperature
Buffer P3 (Neutralisation Buffer)
3.0M potassium acetate, pH 5.5 Stored at 4°C
Buffer QBT (Equilibration Buffer)
750mM NaCl
50mM MOPS (3 [N-Morpholino] propanesulfonic acid), pH 7.0
15% isopropanol
0.15% TritonX-1 GO Stored at room temperature
Buffer QC (Wash Buffer) l.OM NaCl
50mM MOPS, pH 7.0
15% isopropanol Stored at room temperature
Buffer QF (Elution Buffer)
1.25M NaCl
50mM Tris (hydroxymethyl) methyl ammonium chloride, pH 8.5
15% isopropanol Stored at room temperature
231 Buffer QN (Elution Buffer)
L6M NaCl
50mM MOPS, pH7.0
15% isopropanol Stored at room temperature
Buffers containing MOPS or isopropanol were sterilised by filtration.
Optimal Restriction Digest Conditions
Enzyme Recognition Sequence Optimum Temperature Optimum Buffer
Bam HI G >1< GATC 3TC React 3^
S ail G iT C G C 3TC React 10^
Stock solutions of restriction enzymes (10,000Units/ml) and buffers (lOx working solution) were stored at -20®C.
Enzymes were made up to a working solution of lunit/pl in ddH20.
232 Appendix 4-Preparation of Messenger RNA
QuickPrep® M icro mRNA Purification Kit
Oligo (dT) -Cellulose
Oligo (dT)-cellulose at 25mg/ml suspended in a buffer containing 0.15% Kathon CG/ICP
Biocide (registered trademark of Rohm and Haas company).
Extraction Buffer
A buffered aqueous solution of GTC and N-lauroyl sarcosine.
Elution Buffer lOmM Tris (hydroxymethyl) methyl ammonium chloride, pH 7.5
ImM EDTA
High Salt Solution lOmM Tris (hydroxymethyl) methyl ammonium chloride, pH 7.5
ImM EDTA
0.5M NaCl
Low Salt Buffer lOmM Tris (hydroxymethyl) methyl ammonium chloride, pH 7.5
ImM EDTA
O.IM NaCl
233 Glycogen Solution
5-lOmg/ml glycogen in DEPC-treated water
Potassium Acetate Solution
2.5M Potassium acetate, pH 5.0
DEPC-treated Water
0.1 % Diethyl pyrocarbonate (DEPC)
Allow to stand overnight and then autoclaved.
Polypropylene Microspin™ Columns
Note: The extraction buffer and potassium acetate solution are irritants and should be handled with care.
234 Appendix 5- Reverse Transcriptase PCR
First Strand cDNA Synthesis Kit
Bulk First Strand cDNA Reaction Mix (RT)
Cloned, FPLCpure® Murine Reverse Transcriptase,
RNAguard,
RNase/DNase-Free Bovine serum albumin (BSA),
dATP, dCTP, dGTP and dXTP in aqueous buffer.
Dithiothreitol (DTT) Solution
200mM aqueous solution
pd(N)6 Primer
Random hexanucleotides at 0.2pg/pl in aqueous solution.
Plasmid Positive Control Reaction Mix
0.2pg purified plasmid DNA (stored as stock solution 1 mg/ml)
5 pi 1 Ox Supertaq buffer
5 pi 2mM dNTPs
5pi 1 OpM stock Primer P 1
5 pi 10pM stock Primer P2
2.5 units Supertaq
Made up to 50pl with ddH20
235 Optimal PCR Reaction Conditions For Primers PI and P2
Step 1 94°C 5 minutes xl Initial dénaturation
Step 2 94°C 1 minute Dénaturation
5 9 T 2 minutes x35 Annealing
72°C 2 minutes Extension
Step 3 72T 10 minutes xl Final extension
Optimal PCR Reaction Conditions For Primers MBAF and MBAR
The basic protocol was as above, but tbe annealing temperature was increased to 61°C and tbe cycle number was decreased to 33.
236 Appendix 6-Preparation of The Protein Sample
Complete^’^ Protease Inhibitor Cocktail Tablets
Complete™ contains both reversible and irreversible protease inhibitors, and will inhibit the action of serine, cysteine and metalloproteases. It has been shown to be active against
concentrated pancreatic extracts and purified pronase, thermolysin, chymotrypsin, trypsin
and papain. One Complete^^ tablet was dissolved in 50ml of the appropriate lysis buffer
for immediate used or dissolved in 2ml ddHzO and stored at -4°C as stock solution. Stock
solutions of Complete™ were further diluted in 25 volumes of lysis buffer.
APC Lysis Buffers
'Smith’ Lysis Buffer
63mM Tris (hydroxymethyl) methyl ammonium chloride, pH 6.8
2% SDS
10% Glycerol
1 OOmM Dithiothreitol (DTT) or 5% p-mercaptoethanol
0.025% Bromophenol Blue (added after protein quantitation)
DTT maintains -SH groups in a reduced state. It is added immediately prior to use from
IM stock stored at -20°C.
'Chop' Lysis Buffer
0.5M Tris (hydroxymethyl) aminomethane, pH6.8
10% SDS
10% Glycerol
237 0.5% p-mercaptoethanol
0.05% Bromophenol Blue
66.7mM Phenylmethylsulfonylfluoride (PMSF)
0.33mM EDTA
PMSF is an inhibitor of serine proteases. It is extremely destructive to mucous membranes. Contact with eyes, skin or inhalation should be avoided. It can be safely discarded if rendered alkaline (pH>8.6) and stored for several hours at room temperature.
P-Catenin Lysis Buffers
’Alman' Lysis Buffer
lOmM Tris (hydroxymethyl) methyl ammonium chloride, pH 7.4
1% SDS
P-Catenin Loading Buffer
IM Tris (hydroxymethyl) aminomethane, pH 6.8
10% SDS
10% Glycerol
10% p-mercaptoethano 1
0.25% Bromophenol Blue
BCA Protein Assay Reagents
Reagant A is a solution containing BCA, sodium bicarbonate, sodium carbonate, and
sodium tartrate in O.IM sodium hydroxide. Reagent B is a solution of 4% cupric sulphate,
a source of Cu^^ ions.
238 In the alkaline medium produced when the reagents are mixed together, protein present in solution reduces Cu^"^ to Cu^^. One molecule of Cu^^ chelates with two molecules of BCA to give a purple reaction product that absorbs light with an incident wavelength of 562nm.
The amount of light absorbed increases linearly with increasing protein concentration from 20 pg/ml to 2,0000 pg/ml.
239 Appendix 7-Denaturing Gel Electrophoresis
TBE/0.1% SDS Solution For Denaturing Agarose Gel
89mM Tris (hydroxymethyl) aminomethane
89mM Boric acid
2mM EDTA, pH 8
Made up to 11 in ddHiO and autoclaved. 10ml of 10% SDS is then added to 990ml of this solution. It is stored at room temperature. 3g of low melting point agarose is added to
100ml of this solution and heated in a 75OW microwave (on full power) for 2 minutes with intermittent agitation until it dissolves. The gel is then poured while still hot (>65°C) into a horizontal gel tray and the combs added.
Tris-Glycine Running Buffer
25mM Tris (hydroxymethyl) methyl ammonium chloride
192mM Glycine
0.1% SDS
Stored at room temperature.
8% Resolving SDS-PAGE Gel
A 20ml volume is adequate for 2 small ATTO gels or 1 large Hoefer gel. 50ml is used to produce 2 Hoefer gels. TEMED (N,N,N’,N'-tetramethylethylenediamine) catalyses the formation of free radicals from APS. Free radicals in turn catalyse the polymerisation of the acrylamide mix.
240 20ml 50ml ddHzO 9.3ml 26.5ml
30% acrylamide mix 5.3ml 10ml
1.5M Tris, pH 8.8 5.0ml 12.5ml
10% SDS 0.2ml 0.5ml
10% Ammonium persulphate (APS) 0.2ml 0.5ml
TEMED 12pl 40pl
3% Stacking SDS-PAGE Gel
A 10ml volume is adequate for 2 ATTO gels or 1 Hoefer gel. 40ml is more than enough to produce 2 Hoefer gels.
10ml 40ml ddHzO 6.8ml 30.14ml
30% acrylamide mix 1.7ml 4.0ml
IM Tris, pH 6.8 1.25ml 5.0ml
10% SDS 0.1ml 0.4ml
10% APS 0.1ml 0.4ml
TEMED lOpl 64pl
241 Appendix 8- Staining of Protein Gels
Coomassie Brilliant Blue Staining
The fixative/ wash solution is made up of
45% Methanol
45% ddHzO
10% Glacial acetic acid
0.25g of Coomassie Brilliant Blue R250 (Sigma-Aldrich Ltd, Poole, England) is added to
100ml of wash solution to produce the staining solution. This is filtered prior to use to remove particulate material.
Silver Staining -Fixative
30% Ethanol
60% ddHzO
10% Glacial acetic acid
242 Appendix 9-Transfer To A Membrane Support
Capillary Transfer lOX Tris Buffered Saline (TBS)
200mM Tris (hydroxymethyl) aminomethane
1.37M NaCl
Made up as ten times stock solution, pH corrected to 7.6 and autoclaved.
Diluted to working concentration (IX) when needed.
Stored at room temperature.
Capillary Transfer Buffer
0.04% SDS in TBS
Mechanism of Action of Capillary Transfer
■ Pre-soaked Whatmann paper
□ Gel
□ Nitrocellulose
■ Dry Whatmann paper
1 Blotting paper
Transfer buffer
243 Electroblotting
Transfer Buffer
25mM Tris (hydroxymethyl) aminomethane
192mM Glycine
20% Methanol
Stored at 4”C
Mechanism of Action of Electroblotting
□ Transfer buffer
L'l Plastic supports
□ Porous pads
■ Pre-soaked Whatmann paper
□ Nitrocellulose membrane
□ Gel
244 Appendix 10- Staining of Membranes
Ponceau S Staining
10 X Ponceau S Stain
2g Ponceau S powder
30g Trichloroacetic acid
30g Sulphosalicylic acid
Made upto 11 with ddHiO.
This stock solution was diluted in ten volumes of ddH20 to make a working solution.
The working solution was discarded after use.
India Ink Staining
Phosphate Buffered Saline (PBS), pH 7.4
This is made up by dissolving 1 PBS 'tablet' (Sigma Aldrich Company Ltd) in 200ml ddHzO.
Soaking Solution
0.4% Tween 20 in PBS
Staining Solution
0.1% India Ink in Soaking Solution
245 Appendix 11-Immunoblotting
TEST
0.1% Tween 20 in TBS
The components of TBS are described in Appendix 11.
Blocking Solution
5% Marvel™ in TBST
Primary Antibodies
Ab-1
Raised against an epitope in the N-terminal 35 amino acids of APC (Oncogene Research
Products, Cambridge, Massachusetts, USA).
Stored at 4°C. Used at 1/100 final dilution.
C l9220
Raised against an epitope in the C-terminal 180 amino acids of P-catenin (Transduction
Laboratories, Lexington, Kentucky, USA).
Stored at -20°C. Used at 1/2000 final dilution.
A5441
Raised against a synthetic N-terminal 16 amino acid peptide of P-actin (Sigma-Aldrich,
Poole, Dorset, UK).
Stored at -20°C. Used at 1/5000 final dilution
246 C1801
Raised against a cytokeratin preparation from human epidermis made up of cytokeratins
1, 5, 6 and 8. Cytokeratin 8 (52kD) is a major type II keratin in simple epithelia such as the intestine. Purchased from Sigma-Aldrich, Poole, Dorset, UK.
Stored at -20®C in separate aliquots. Used at 1/500 final dilution.
Strip Buffer
0.02M glycine, pH 2.5
0.05% Triton X-100
247 Appendix 12- Immunohistochemistry
Mechanism of Action of Immunohistochemistry
Secondary Ab □ Antigen
Primary Ab / ■ Biotin
Streptavidin Tissue Section ■ HRPO
Haematoxylin and Eosin Staining of Tissue Sections
Dip sections in haematoxylin solution for 5-7 minutes
Rinse briefly under running water
Destain in 0.1% acid-alcohol for 5 seconds
Rinse well under running water until only the nuclei are stained
'Blue' the haematoxylin in Scott's tap water for 1 minute
Wash under running water for 10 minutes
Stain in 0.1% eosin for 5 minutes
Wash briefly in running water
Dehydrate in a series of alcohols (70%, 100%, 100%) and at least 2 pots of xylene for
1 minute each
Mount in diphenylxylene with a coverslip
248 Haematoxylin Solution
7.5g Haematoxylin monohydrate
150g Ammonium or potassium alum
4.5g Yellow mercuric oxide
8ml Glacial acetic acid
Make upto 2 1 with ddHzO
Dissolve the alum in warm distilled water, add the haematoxylin and bring to the boil.
Allow to cool in cold water and add mercuric oxide and glacial acetic acid.
0.1% Acid-Alcohol
99ml 70% alcohol
1 ml concentrated HCl
Scott's Tap Water
7g Sodium bicarbonate
40g Magnesium sulphate
Make up to 2 1 with tap water
249 Chapter 9
References
250 Aaltonen, L. A., Peltomaki, P., Leach, F. S., Sistonen, P., Pylkkanen, L., Mecklin, J. P.,
Jarvinen, H., Powell, S. M., Jen, L, Hamilton, S. R. and et al. (1993) Clues to the
pathogenesis of familial colorectal cancer [see comments]. Science, 260, 812-6.
Aarnio, M., Mustonen, H., Mecklin, J. P. and Jarvinen, H. J. (1998) Prognosis of
colorectal cancer varies in different high-risk conditions, Ann Med, 30, 75-80.
Aberle, H., Bauer, A., Stappert, J., Kispert, A. and Kemler, R. (1997) beta-catenin is a
target for the ubiquitin-proteasome pathway, Embo J, 16, 3797-804.
Aksentijevich, I., Pastan, I., Lunardi-Iskandar, Y., Gallo, R. C., Gottesman, M. M. and
Thierry, A. R. (1996) In vitro and in vivo liposome-mediated gene transfer leads
to human MDRl expression in mouse bone marrow progenitor cells. Hum Gene
Ther, 7, 1111-22.
Allay, J. A., Persons, D. A., Galipeau, J., Riberdy, J. M., Ashmun, R. A., Blakley, R. L.
and Sorrentino, B. P. (1998) In vivo selection of retrovirally transduced
hematopoietic stem cells, Nat Med, 4, 1136-43.
Alman, B. A., Li, C., Pajerski, M. E., Diaz-Cano, S. and Wolfe, H. J. (1997) Increased
beta-catenin protein and somatic APC mutations in sporadic aggressive
fibromatoses (desmoid tumors). Am J Pathol, 151, 329-34.
Anwer, K., Earle, K. A., Shi, M., Wang, J., Mumper, R. J., Proctor, B., Jansa, K.,
Ledebur, H. C., Davis, S., Eaglstein, W. and Rolland, A. P. (1999) Synergistic
effect of formulated plasmid and needle-free injection for genetic vaccines, Pharm
Res, 16, 889-95.
Aoki, K., Yoshida, T., Matsumoto, N., Ide, H., Hosokawa, K., Sugimura, T. and Terada,
M. (1997) Gene therapy for peritoneal dissemination of pancreatic cancer by
liposome-mediated transfer of herpes simplex virus thymidine kinase gene. Hum
Gene Ther, 8, 1105-13.
251 Arber, N., Hibshoosh, H., Moss, S. F., Sutter, T., Zhang, Y., Begg, M., Wang, S.,
Weinstein, I. B. and Holt, P. R. (1996) Increased expression of cyclin D1 is an
early event in multistage colorectal carcinogenesis. Gastroenterology, 110, 669-
74.
Arenas, R. B., Fichera, A., Mok, P., Blanco, M. C. and Michelassi, F. (1996) Introduction
of human adenomatous polyposis coli gene into Min mice via cationic liposomes.
Surgery, 120, 712-7; discussion 717-8.
Aruffo, A., Stamenkovic, I., Melnick, M., Underhill, C. B. and Seed, B. (1990) CD44 is
the principal cell surface receptor for hyaluronate. Cell, 61, 1303-13.
Arvanitis, M. L., Jagelman, D. G., Fazio, V. W., Lavery, I. C. and McGannon, E. (1990)
Mortality in patients with familial adenomatous polyposis. Dis Colon Rectum, 33,
639-42.
Asgari, K., Sesterhenn, I. A., McLeod, D. G., Cowan, K., Moul, J. W., Seth, P. and
Srivastava, S. (1997) Inhibition of the growth of pre-established subcutaneous
tumor nodules of human prostate cancer cells by single injection of the
recombinant adenovirus p53 expression vector, Int J Cancer, 71, 377-82.
Baeg, G. H., Matsumine, A., Kuroda, T., Bhattacharjee, R. N., Miyashiro, I., Toyoshima,
K. and Akiyama, T. (1995) The tumour suppressor gene product APC blocks cell
cycle progression from GO/Gl to S phase, Embo J, 14, 5618-25.
Baker, S. J., Markowitz, S., Fearon, E. R., Willson, J. K. and Vogelstein, B. (1990)
Suppression of human colorectal carcinoma cell growth by wild-type p53.
Science, 249, 912-5.
Ballo, M. T., Zagars, G. K., Pollack, A., Pisters, P. W. and Pollack, R. A. (1999) Desmoid
tumor: prognostic factors and outcome after surgery, radiation therapy, or
combined surgery and radiation therapy, J C/zw Oncol, 17, 158-67.
252 Bames, C. J. and Lee, M. (1998) Chemoprevention of spontaneous intestinal adenomas in
the adenomatous polyposis coli Min mouse model with aspirin. Gastroenterology,
114, 873-7.
Beach, D., Piper, M. and Shall, S. (1980) Isolation of chromosomal origins of replication
in yeast. Nature, 284, 185-7.
Beazer-Barclay, Y., Levy, D. B., Moser, A. R., Dove, W. F., Hamilton, S. R., Vogelstein,
B. and Kinzler, K. W. (1996) Sulindac suppresses tumorigenesis in the Min
mouse. Carcinogenesis, 17, 1757-60.
Behrens, J., Vakaet, L., Friis, R., Winterhager, E., Van Roy, F., Mareel, M. M. and
Birchmeier, W. (1993) Loss of epithelial differentiation and gain of invasiveness
correlates with tyrosine phosphorylation of the E-cadherin/beta-catenin complex
in cells transformed with a temperature-sensitive v-SRC gene, J Cell Biol, 120,
757-66.
Behrens, J., von Kries, J. P., Kuhl, M., Bruhn, L., Wedlich, D., Grosschedl, R. and
Birchmeier, W. (1996) Functional interaction of beta-catenin with the
transcription factor CEV-\, Nature, 382, 638-42.
Bergers, G., Javaherian, K., Lo, K. M., Folkman, J. and Hanahan, D. (1999) Effects of
angiogenesis inhibitors on multistage carcinogenesis in mice. Science, 284, 808-
12.
Beroud, C. and Soussi, T. (1996) APC gene: database of germline and somatic mutations
in human tumors and cell lines. Nucleic Acids Res, 24, 121-4.
Berrueta, L., Kraeft, S. K., Timauer, J. S., Schuyler, S. C., Chen, L. B., Hill, D. E.,
Pellman, D. and Bierer, B. E. (1998) The adenomatous polyposis coli-binding
protein EBl is associated with cytoplasmic and spindle microtubules, Proc Natl
AcadSci USA, 95, 10596-601.
253 Bertario, L., Russo, A., Sala, P., Eboli, M., Radice, P., Presciuttini, S., Andreola, S.,
Rodriguez-Bigas, M. A., Pizzetti, P. and Spinelli, P. (1999) Survival of patients
with hereditary colorectal cancer: comparison of HNPCC and colorectal cancer in
FAP patients with sporadic colorectal cancer, Int J Cancer, 80, 183-7.
Bhat, R. V., Baraban, J. M., Johnson, R. C., Eipper, B. A. and Mains, R. E. (1994) High
levels of expression of the tumor suppressor gene APC during development of the
rat central nervous system, JNeurosci, 14, 3059-71.
Bi, W. L., Parysek, L. M., Wamick, R. and Stambrook, P. J. (1993) In vitro evidence that
metabolic cooperation is responsible for the bystander effect observed with HSV
tk retroviral gene therapy. Hum Gene Ther, 4, 725-31.
Birnboim, H. C. and Doly, J. (1979) A rapid alkaline extraction procedure for screening
recombinant plasmid DNA, Nucleic Acids Res, 7, 1513-23.
Bishop, J. M. (1991) Molecular themes in oncogenesis. Cell, 64,235-48.
Black, R., Sharp, L. and Kendrick, S. (1993),ISD Publications, Edinburgh.
Blaese, R. M., Culver, K. W., Miller, A. D., Carter, C. S., Fleisher, T., Clerici, M.,
Shearer, G., Chang, L., Chiang, Y., Tolstoshev, P. and et al. (1995) T lymphocyte-
directed gene therapy for ADA- SCID: initial trial results after 4 years. Science,
270, 475-80.
Blaschuk, O. W., Sullivan, R., David, S. and Pouliot, Y. (1990) Identification of a
cadherin cell adhesion recognition sequence, Dev Biol, 139,227-9.
Bloom, K. S. and Carbon, J. (1982) Yeast centromere DNA is in a unique and highly
ordered structure in chromosomes and small circular minichromosomes. Cell, 29,
305-17.
Bocker, T., Schlegel, J., Kullmann, F., Stumm, G., Zirngibl, H., Epplen, J. T. and
Ruschoff, J. (1996) Genomic instability in colorectal carcinomas: comparison of
different evaluation methods and their biological significance, J Pathol, 179, 15-9.
254 Bonneton, C., Larue, L. and Thiery, J. P. (1996) The APC gene product and colorectal
carcinogenesis, C R Acad Sci III, 319, 861-9.
Boolbol, S. K., Dannenberg, A. J., Chadburn, A., Martucci, C., Guo, X. J., Ramonetti, J.
T., Abreu-Goris, M., Newmark, H. L., Lipkin, M. L., DeCosse, J. J. and
Bertagnolli, M. M. (1996) Cyclooxygenase-2 overexpression and tumor formation
are blocked by sulindac in a murine model of familial adenomatous polyposis.
Cancer Res, 56, 2556-60.
Boussif, O., Lezoualc’h, P., Zanta, M. A., Mergny, M. D., Scherman, D., Demeneix, B.
and Behr, J. P. (1995) A versatile vector for gene and oligonucleotide transfer into
cells in culture and in vivo: polyethylenimine, Proc Natl Acad Sci USA, 92,
7297-301.
Bouvet, M., Ellis, L. M., Nishizaki, M., Fujiwara, T., Liu, W., Bucana, C. D., Fang, B.,
Lee, J. J. and Roth, J. A. (1998a) Adenovirus-mediated wild-type p53 gene
transfer down-regulates vascular endothelial growth factor expression and inhibits
angiogenesis in human colon cancer. Cancer Res, 58, 2288-92.
Bouvet, M., Fang, B., Ekmekcioglu, S., Ji, L., Bucana, C. D., Hamada, K., Grimm, E. A.
and Roth, J. A. (1998b) Suppression of the immune response to an adenovirus
vector and enhancement of intratumoral transgene expression by low-dose
etoposide. Gene Ther, 5, 189-95.
Boyle, P. and Langman, J. S. (2000) Epidemiology, Bmj, 321, 805-8.
Brand, K., Loser, P., Arnold, W., Bartels, T. and Strauss, M. (1998) Tumor cell-specific
transgene expression prevents liver toxicity of the adeno-HSVtk/GCV approach.
Gene Ther, 5, 1363-71.
Brensinger, J. D., Taken, S. J., Luce, M. C., Powell, S. M., Vance, G. H., Ahnen, D. J.,
Petersen, G. M., Hamilton, S. R. and Giardiello, F. M. (1998) Variable phenotype
255 of familial adenomatous polyposis in pedigrees with 3' mutation in the APC gene
[see comments], Gut, 43, 548-52.
Bulow, S., Faurschou Nielsen, T., Bulow, C., Bisgaard, M. L., Karlsen, L. and
Moesgaard, F. (1996) The incidence rate of familial adenomatous polyposis.
Results from the Danish Polyposis Register, Int J Colorectal Dis, 11, 88-91.
Burnette, W. N. (1981) "Western blotting": electrophoretic transfer of proteins from
sodium dodecyl sulfate—polyacrylamide gels to unmodified nitrocellulose and
radiographic detection with antibody and radioiodinated protein A, Anal Biochem,
112, 195-203.
Burt, R. W., DiSario, J. A. and Cannon-Albright, L. (1995) Genetics of colon cancer:
impact of inheritance on colon cancer risk, Annu Rev Med, 46, 371-9.
Bussey, H. (1975) Familail Polyposis Coli: Family Studies, Histopathology, Differential
Diagnosis, and Results of Treatment, Johns Hopkins University Press, Baltimore,
Md.
Campbell, W. J., Spence, R. A. and Parks, T. G. (1994) Familial adenomatous polyposis,
B rJSurg, 81, 1722-33.
Cao, X., Eu, K. W., Seow-Choen, F. and Cheah, P. Y. (1999) Germline mutations are
frequent in the APC gene but absent in the beta- catenin gene in familial
adenomatous polyposis patients. Genes Chromosomes Cancer, 25, 396-8.
Caplen, N. J., Alton, E. W., Middleton, P. G., Dorin, J. R., Stevenson, B. J., Gao, X.,
Durham, S. R., Jeffery, P. K., Hodson, M. E., Coutelle, C. and et al. (1995a)
Liposome-mediated CFTR gene transfer to the nasal epithelium of patients with
cystic fibrosis [see comments] [published erratum appears in Nat Med 1995
Mar;l(3):272], Nat Med, 1, 39-46.
256 Caplen, N. J., Higginbotham, J. N., Scheel, J. R., Vahanian, N., Yoshida, Y., Hamada, H.,
Blaese, R. M. and Ramsey, W. J. (1999) Adeno-retroviral chimeric viruses as in
vivo transducing agents. Gene Ther, 6,454-9.
Caplen, N. J., Kinrade, E., Sorgi, F., Gao, X., Gruenert, D., Geddes, D., Coutelle, C.,
Huang, L., Alton, E. W. and Williamson, R. (1995b) In vitro liposome-mediated
DNA transfection of epithelial cell lines using the cationic liposome DC-
Cho 1/DOPE, Gene Ther, 2, 603-13.
Cartwright, C. A., Meisler, A. I. and Eckhart, W. (1990) Activation of the pp60c-src
protein kinase is an early event in colonic carcinogenesis, Proc Natl Acad Sci U S
A, 87, 558-62.
Caspari, R., Friedl, W., Mandl, M., Moslem, G., Kadmon, M., Knapp, M., Jacobasch, K.
H., Ecker, K. W., Kreissler-Haag, D., Timmermanns, G. and et al. (1994) Familial
adenomatous polyposis: mutation at codon 1309 and early onset of colon cancer
[published erratum appears in Lancet 1994 Apr 2;343(8901):863], Lancet, 343,
629-32.
Caspari, R., Olschwang, S., Friedl, W., Mandl, M., Boisson, C., Boker, T., Augustin, A.,
Kadmon, M., Moslem, G., Thomas, G. and et al. (1995) Familial adenomatous
polyposis: desmoid tumours and lack of ophthalmic lesions (CHRPE) associated
with APC mutations beyond codon 1444, Hum Mol Genet, 4, 337-40.
Chen, P. L., Chen, Y. M., Bookstein, R. and Lee, W. H. (1990) Genetic mechanisms of
tumor suppression by the human p53 gene. Science, 250, 1576-80.
Chien, A., Edgar, D. B. and Trela, J. M. (1976) Deoxyribonucleic acid polymerase from
the extreme thermophile Thermus aquaticus,JBacteriol, 127, 1550-7.
Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J. and Rutter, W. J. (1979) Isolation of
biologically active ribonucleic acid from sources enriched in ribonuclease.
Biochemistry, 18, 5294-9.
257 Chirmule, N., Propert, K., Magosin, S., Qian, Y., Qian, R. and Wilson, J. (1999) Immune
responses to adenovirus and adeno-associated virus in humans [In Process
Citation], Gene Ther, 6, 1574-83.
Chop, A. M., Abraham, C. L., Adrian, T. E., Murphy, R. and Boman, B. M. (1995)
Immunodetection of the presence or absence of full-length APC gene product in
human colonic tissues. Anticancer Res, 15, 991-7.
Clark, S. K., Neale, K. P., Landgrebe, J. C. and Phillips, R. K. (1999) Desmoid tumours
complicating familial adenomatous polyposis, B rJSurg, 86, 1185-9.
Clark, S. K. and Phillips, R. K. (1996) Desmoids in familial adenomatous polyposis [see
comments], 83, 1494-504. dayman, G. L., el-Naggar, A. K., Lippman, S. M., Henderson, Y. C., Frederick, M.,
Merritt, J. A., Zumstein, L. A., Timmons, T. M., Liu, T. J., Ginsberg, L., Roth, J.
A., Hong, W. K., Bruso, P. and Goepfert, H. (1998) Adenovirus-mediated p53
gene transfer in patients with advanced recurrent head and neck squamous cell
carcinoma, J C/m Oncol, 16, 2221-32. daym an, G. L., el-Naggar, A. K., Roth, J. A., Zhang, W. W., Goepfert, H., Taylor, D. L.
and Liu, T. J. (1995) In vivo molecular therapy with p53 adenovirus for
microscopic residual head and neck squamous carcinoma. Cancer Res, 55, 1-6.
Cline, M. J., Stang, H., Mercola, K., Morse, L., Ruprecht, R., Brown, J. and Salser, W.
(1980) Gene transfer in intact animals. Nature, 284, 422-5.
Cole-Strauss, A., Yoon, K., Xiang, Y., Byrne, B. C., Rice, M. C., Gryn, J., Holloman, W.
K. and Kmiec, E. B. (1996) Correction of the mutation responsible for sickle cell
anemia by an RNA- DNA oligonucleotide [see comments]. Science, 273, 1386-9.
Conry, R. M., LoBuglio, A. P., Loechel, P., Moore, S. E., Sumerel, L. A., Barlow, D. L.,
Pike, J. and Curiel, D. T. (1995) A carcinoembryonic antigen polynucleotide
vaccine for human clinical use. Cancer Gene Ther, 2, 33-8.
258 Conry, R. M., Strong, T., White, S., Khazaeli, M., LoBuglio, A. F. and Curiel, D. T.
(1997) Cancer Gene Therapy, 4, S49.
Conry, R. M., Widera, G., LoBuglio, A. F., Fuller, J. T., Moore, S. E., Barlow, D. L.,
Turner, L, Yang, N. S. and Curiel, D. T. (1996) Selected strategies to augment
polynucleotide immunization. Gene Ther, 3, 67-74.
Cormier, R. T., Hong, K. H., Halberg, R. B., Hawkins, T. L., Richardson, P., Mulherkar,
R., Dove, W. F. and Lander, E. S. (1997) Secretory phospholipase Pla2g2a
confers resistance to intestinal tumorigenesis [see comments], Nat Genet, 17, 88-
91.
Cotter, F. E., Johnson, P., Hall, P., Pocock, C., al Mahdi, N., Cowell, J. K. and Morgan,
G. (1994) Antisense oligonucleotides suppress B-cell lymphoma growth in a
SCID-hu mouse model. Oncogene, 9, 3049-55.
Crystal, R. G., Hirschowitz, E., Lieberman, M., Daly, J., Kazam, E., Henschke, C.,
Yankelevitz, D., Kemeny, N., Silverstein, R., Ohwada, A., Russi, T., Mastrangeli,
A., Sanders, A., Cooke, J. and Harvey, B. G. (1997) Phase I study of direct
administration of a replication deficient adenovirus vector containing the E. coli
cytosine deaminase gene to metastatic colon carcinoma of the liver in association
with the oral administration of the pro-drug 5-fiuorocytosine, Hum Gene Ther, 8,
985-1001.
D'Abaco, G. M., Whitehead, R. H. and Burgess, A. W. (1996) Synergy between Ape min
and an activated ras mutation is sufficient to induce colon carcinomas. Mol Cell
Biol, 16, 884-91.
Dangel, A., Meloni, A. M., Lynch, H. T. and Sandberg, A. A. (1994) Deletion (5q) in a
desmoid tumor of a patient with Gardner's syndrome. Cancer Genet Cytogenet,
78, 94-8.
259 Davies, D. R., Armstrong, J. G., Thakker, N., Homer, K., Guy, S. P., Clancy, T., Sloan,
P., Blair, V., Dodd, C., Wames, T. W. and et al. (1995) Severe Gardner syndrome
in families with mutations restricted to a specific region of the APC gene. Am J
Hum Genet, 57, 1151-8.
Davis, B. J. (1964) Disc electrophoresis-II. Method and application to human serum
proteins., Ann. N. Y Acad. Sci, 121, 404-. de Wind, N., Dekker, M., Claij, N., Jansen, L., van Klink, Y., Radman, M., Riggins, G.,
van der Valk, M., van't Wout, K. and te Riele, H. (1999) HNPCC-like cancer
predisposition in mice through simultaneous loss of Msh3 and Msh6 mismatch-
repair protein functions, Nat Genet, 23, 359-62.
Delatycki, M., Nasioulas, S., Forshaw, K. and Forrest, S. (1998) Two novel mutations in
exons 5 and 15 of the adenomatous polyposis coli (APC) gene. Hum Mutat,
Suppl, S314-6.
Devereux, S., Comey, C., Macdonald, C., Watts, M., Sullivan, A., Goldstone, A. H.,
Ward, M., Bank, A. and Linch, D. C. (1998) Feasibility of multidrug resistance
(MDR-1) gene transfer in patients undergoing high-dose therapy and peripheral
blood stem cell transplantation for lymphoma. Gene Ther, 5, 403-8.
Dietrich, W. F., Lander, E. S., Smith, J. S., Moser, A. R., Gould, K. A., Luongo, C.,
Borenstein, N. and Dove, W. (1993) Genetic identification of Mom-1, a major
modifier locus affecting Min- induced intestinal neoplasia in the mouse. Cell, 75,
631-9.
Dihlmann, S., Amler, L. C., Schwab, M. and Wenzel, A. (1997) Variations in the
expression of the adenomatous polyposis coli (APC) tumor suppressor gene in
human cancer cell lines of different tissue origin, Oncol Res, 9, 119-27.
260 Dilber, M. S., Phelan, A., Aints, A., Mohamed, A. J., Elliott, G., Edvard Smith, C. I. and
O'Hare, P. (1999) Intercellular delivery of thymidine kinase prodrug activating
enzyme by the herpes simplex virus protein, VP22, Gene Ther, 6, 12-21.
Dion, L. D., Fang, J. and Garver, R. L, Jr. (1996) Supernatant rescue assay vs. polymerase
chain reaction for detection of wild type adenovirus-contaminating recombinant
adenovirus stocks, J Virol Methods, 56, 99-107.
Dobbie, Z. and Muller, H. (1999) Germline mutations in the beta-eatenin gene are not
associated with the FAP phenotype without an APC mutation [letter], J Med
Genet, 36, 573-4.
DuBois, R. N. and Smalley, W. E. (1996) Cyelooxygenase, NSAIDs, and colorectal
cancer, J Gastroenterol, 31, 898-906.
During, M., Xu, R., Young, D., Kaplitt, M., Sherwin, R. and Leone, P. (1998) Peroral
gene therapy of lactose intolerance using an adeno-associated virus vector. Nature
Medicine, 4, 1131-1135.
Duval, A., Gayet, J., Zhou, X., laeopetta, B., Thomas, G. and Hamelin, R. (1999)
Frequent frameshifl mutations of the TCF-4 gene in colorectal cancers with
mierosatellite instability.. Cancer Research, 59, 4213-4215.
Eberhart, C. E., Coffey, R. J., Radhika, A., Giardiello, F. M., Ferrenbaeh, S. and DuBois,
R. N. (1994) Up-regulation of cyelooxygenase 2 gene expression in human
colorectal adenomas and adenocarcinomas. Gastroenterology, 107, 1183-8.
Einstein, D. M., Tagliabue, J. R. and Desai, R. K. (1991) Abdominal desmoids: CT
findings in 25 patients, AJR Am J Roentgenol, 157, 275-9.
Eldar-Finkelman, H., Seger, R., Vandenheede, J. R. and Krebs, E. G. (1995) Inactivation
of glycogen synthase kinase-3 by epidermal growth factor is mediated by
mitogen-activated protein kinase/p90 ribosomal protein S6 kinase signaling
pathway in NIH/3T3 cells, J Biol Chem, 270, 987-90.
261 Elliott, G. and O'Hare, P. (1997) Intercellular trafficking and protein delivery by a
herpesvirus structural protein. Cell, 88, 223-33.
Elliott, G. and O'Hare, P. (1999) Intercellular trafficking of VP22-GFP fusion proteins.
Gene Ther, 6, 149-51.
Elliott, G. D. and Meredith, D. M. (1992) The herpes simplex virus type 1 tegument
protein VP22 is encoded by gene UL49, J Gen Virol, 73, 723-6.
Fearon, E. R. and Vogelstein, B. (1990) A genetic model for colorectal tumorigenesis.
Cell, 61, 759-67.
Feigner, P. L., Gadek, T. R., Holm, M., Roman, R , Chan, H. W., Wenz, M., Northrop, J.
P., Ringold, G. M. and Danielsen, M. (1987) Lipofection: a highly efficient, lipid-
mediated DNA-transfection procedure, Proc Natl Acad Sci USA, 84, 7413-7.
Fichera, A., Guo, Y., Romero, L., Michelassi, F. and Arenas, R. B. (1997) Quantitation of
in vivo gene delivery by restriction enzyme PGR generated polymorphism, J Surg
Res, 69, 188-92.
Fishel, R., Lescoe, M. K., Rao, M. R., Copeland, N. G., Jenkins, N. A., Garber, J., Kane,
M. and Kolodner, R. (1993) The human mutator gene homolog MSH2 and its
association with hereditary nonpolyposis colon cancer [published erratum appears
in Cell 1994 Apr 8;77(1):167], Cell, 75, 1027-38.
Fodde, R., Edelmann, W., Yang, K., van Leeuwen, C., Carlson, C., Renault, B., Breukel,
C., Alt, E., Lipkin, M., Khan, P. M. and et al. (1994) A targeted chain-termination
mutation in the mouse Ape gene results in multiple intestinal tumors, Proc Natl
A cadSci USA, 91, 8969-73.
Fodde, R., van der Luijt, R., Wijnen, J., Tops, C., van der Klift, H., van Leeuwen-
Comelisse, I., Griffioen, G., Vasen, H. and Klian, P. M. (1992) Eight novel
inactivating germ line mutations at the APC gene identified by denaturing
gradient gel electrophoresis. Genomics, 13, 1162-8.
262 Folkman, J. (1990) What is the evidence that tumors are angiogenesis dependent?
[editorial], J Natl Cancer Inst, 82, 4-6.
Franke, W. W., Goldschmidt, M. D., Zimbelmann, R., Mueller, H. M., Schiller, D. L. and
Co win, P. (1989) Molecular cloning and amino acid sequence of human
plakoglobin, the common junctional plaque protein, Proc Natl Acad Sci USA, 86,
4027-31.
Frayling, I. M., Beck, N. E., Ilyas, M., Dove-Edwin, I., Goodman, P., Pack, K., Bell, J.
A., Williams, C. B., Hodgson, S. V., Thomas, H. J., Talbot, I. C., Bodmer, W. F.
and Tomlinson, I. P. (1998) The APC variants I1307K and E1317Q are associated
with colorectal tumors, but not always with a family history, Proc Natl Acad Sci U
SA, 95,10722-7.
Friedl, W., Meuschel, S., Caspari, R., Lamberti, C., Krieger, S., Sengteller, M. and
Propping, P. (1996) Attenuated familial adenomatous polyposis due to a mutation
in the 3' part of the APC gene. A clue for understanding the function of the APC
protein. Hum Genet, 97, 579-84.
Fujiwara, T., Cai, D. W., Georges, R. N., Mukhopadhyay, T., Grimm, E. A. and Roth, J.
A. (1994a) Therapeutic effect of a retroviral wild-type p53 expression vector in an
orthotopic lung cancer model [see comments], J Natl Cancer Inst, 86, 1458-62.
Fujiwara, T., Grimm, E. A., Mukhopadhyay, T., Zhang, W. W., Owen-Schaub, L. B. and
Roth, J. A. (1994b) Induction of chemosensitivity in human lung cancer cells in
vivo by adenovirus-mediated transfer of the wild-type p53 gene. Cancer Res, 54,
2287-91.
Funayama, N., Fagotto, F., McCrea, P. and Gumbiner, B. M. (1995) Embryonic axis
induction by the armadillo repeat domain of beta- catenin: evidence for
intracellular signaling, J Ce// Biol, 128, 959-68.
263 Galderisi, U., Cascino, A. and Giordano, A. (1999) Antisense oligonucleotides as
therapeutic agents, J Ce// Physiol, 181, 251-7.
Gao, X. and Huang, L. (1996) Potentiation of cationic liposome-mediated gene delivery
by polycations. Biochemistry, 35, 1027-36.
Gardner, E. and Plenk, H. (1952) Hereditary pattern for multiple osteomas in a family
gvowp., American Journal o f Human Genetics, 4, 31-36.
Gardner, E. and RC, R. (1953) Multiple cutaneous and sub-cutaneous lesions occurring
simultaneously with hereditary polyposis and osteomatosis., American Journal of
Human Genetics, 5, 139-147.
Gardner, R. J., Kool, D., Edkins, E., Walpole, I. R., Macrae, F. A., Nasioulas, S. and
Scott, W. J. (1997) The clinical correlates of a 3' truncating mutation (codons
1982-1983) in the adenomatous polyposis coli gene. Gastroenterology, 113, 326-
31.
Ghadimi, B. M., Behrens, J., Hofhnann, I., Haensch, W., Birchmeier, W. and Schlag, P.
M. (1999) Immunohistological analysis of E-cadherin, alpha-, beta- and gamma-
catenin expression in colorectal cancer: implications for cell adhesion and
signaling, Eur J Cancer, 35, 60-5.
Giardiello, F., Offerhaus, J. and Krush, A. e. a., " (1992) Risk of hepatoblastoma in
familial adenomatous polyposis., Jpaediatr, 119, 766-768.
Giardiello, F. M., Brensinger, J. D., Luce, M. C., Petersen, G. M., Cayouette, M. C.,
Krush, A. J., Bacon, J. A., Booker, S. V., Bufill, J. A. and Hamilton, S. R. (1997)
Phenotypic expression of disease in families that have mutations in the 5' region
of the adenomatous polyposis coli gene, Ann Intern Med, 126, 514-9.
Giardiello, F. M., Hamilton, S. R., Krush, A. J., Piantadosi, S., Hylind, L. M., Celano, P.,
Booker, S. V., Robinson, C. R. and Offerhaus, G. J. (1993) Treatment of colonic
264 and rectal adenomas with sulindac in familial adenomatous polyposis, N Engl J
Med, 328, 1313-6.
Gilly, F. N., Sayag-Beaujard, A. C., Bienvenu, J., Aymard, M., Touraine-Moulin, F.,
Favrot, M., Trillet-Lenoir, V., Pavirani, A., Courtney, M., Francois, Y., Pellet, O.,
Gaillard, C., Glehen, O., Banssillon, V., Vignal, J. and Ross, M. (1998) Gene
therapy with AdV-IL2 (TG 1021) in unresectable digestive adenocarcinoma.
Phase I-II study, first inclusions, Adv Exp Med Biol, 451, 527-30.
Gjerset, R. A., Turla, S. T., Sobol, R. E., Scalise, J. J., Mercola, D., Collins, H. and
Hopkins, P. J. (1995) Use of wild-type p53 to achieve complete treatment
sensitization of tumor cells expressing endogenous mutant p53. Mol Carcinog, 14,
275-85.
Gladieff, L., Chatelut, E., Gaspard, M. H., Skaf, R., de Fomi, M., Mihura, J., Canal, P.
and Bugat, R. (1999) [Intraperitoneal high dose chemotherapy as consolidation
treatment for advanced ovarian carcinoma: a pilot study]. Bull Cancer, 86, 673-7.
Griesenbach, U., Chonn, A., Cassady, R., Hannam, V., Ackerley, C., Post, M., Tanswell,
A. K., Olek, K., O'Brodovich, H. and Tsui, L. C. (1998) Comparison between
intratracheal and intravenous administration of liposome-DNA complexes for
cystic fibrosis lung gene therapy. Gene Ther, 5, 181-8.
Groden, J., Thliveris, A., Samowitz, W., Carlson, M., Gelbert, L., Albertsen, H., Joslyn,
G., Stevens, J., Spirio, L., Robertson, M. and et al. (1991) Identification and
characterization of the familial adenomatous polyposis coli gene. Cell, 66, 589-
600.
Gryfe, R., Di Nicola, N., Gallinger, S. and Redston, M. (1998) Somatic instability of the
APC I1307K allele in colorectal neoplasia [see comments]. Cancer Res, 58, 4040-
3.
265 Gumbiner, B. M. (1995) Signal transduction of beta-catenin, Curr Opin Cell Biol, 7, 634-
40.
Gura, T. (1999) Repairing the genome's spelling mistakes [news]. Science, 285, 316-8.
Gurbuz, A. K., Giardiello, F. M., Petersen, G. M., Krush, A. J., Offerhaus, G. J., Booker,
S. V., Kerr, M. C. and Hamilton, S. R. (1994) Desmoid tumours in familial
adenomatous polyposis. Gut, 35, 377-81.
Haisma, H. J., Pinedo, H. M., Rijswijk, A., der Meulen-Muileman, I., Sosnowski, B. A.,
Ying, W., Beusechem, V. W., Tillman, B. W., Gerritsen, W. R. and Curiel, D. T.
(1999) Tumor-specific gene transfer via an adenoviral vector targeted to the pan
carcinoma antigen EpC AM, Gene Ther, 6,1469-74.
Hanania, E. G., Fu, S., Roninson, I., Zu, Z. and Deisseroth, A. B. (1995) Resistance to
taxol chemotherapy produced in mouse marrow cells by safety-modified
retroviruses containing a human MDR-1 transcription unit [see comments]
[published erratum appears in Gene Ther 1995 Aug;2(6):433], Gene Ther, 2, 279-
84.
Hanif, R., Pittas, A., Feng, Y., Koutsos, M. I., Qiao, L., Staiano-Coico, L., Shiff, S. I. and
Rigas, B. (1996) Effects of nonsteroidal anti-inflammatory drugs on proliferation
and on induction of apoptosis in colon cancer cells by a prostaglandin-
independent pathway, Biochem Pharmacol, 52, 237-45.
Hao, X., Tomlinson, I., Ilyas, M., Palazzo, J. P. and Talbot, I. C. (1997) Reciprocity
between membranous and nuclear expression of beta-catenin in colorectal
tumours, Virchows Arch, 431, 167-72.
Hargest, R. (1996) In Department o f Biochemistry and Molecular GeneticsGriwQxsiXy o f
London, London, pp. 249.
266 Hargest, R., Eldin, A. and Williamson, R. (1998) Gene therapy for familial adenomatous
polyposis. Prolonged expression of the adenomatous polyposis coli gene after
lipofection into mouse colon in vivo, Adv Exp Med Biol, 451, 385-91.
Hargest, R. and Williamson, R. (1995) Expression of the APC gene after transfection into
a colonic cancer cell line. Gut, 37, 826-9.
Harlow, E. and Lane, D. {\9%%) Antibodies:A Laboratory Journal.
Harrington, J. J., Van Bokkelen, G., Mays, R. W., Gustashaw, K. and Willard, H. F.
(1997) Formation of de novo centromeres and construction of first-generation
human artificial microchromosomes [see comments], Nat Genet, 15, 345-55.
Harris, J. D., Gutierrez, A. A., Hurst, H. C., Sikora, K. and Lemoine, N. R. (1994) Gene
therapy for cancer using tumour-specific prodrug activation. Gene Ther, 1, 170-5.
Hatzfeld, M. (1999) In International Review o f Cytology, Vol. 186, pp. 179-224.
He, T. C., Chan, T. A., Vogelstein, B. and Kinzler, K. W. (1999) PPARdelta is an APC-
regulated target of nonsteroidal anti-inflammatory drugs. Cell, 99, 335-45.
He, T. C., Sparks, A. B., Rago, C., Hermeking, H., Zawel, L., da Costa, L. T., Morin, P.
J., Vogelstein, B. and Kinzler, K. W. (1998) Identification of c-MYC as a target of
the APC pathway [see comments]. Science, 281, 1509-12.
Heasman, J., Crawford, A., Goldstone, K., Gamer-Hamrick, P., Gumbiner, B., McCrea,
P., Kintner, C., Noro, C. Y. and Wylie, C. (1994) Overexpression of cadherins and
underexpression of beta-catenin inhibit dorsal mesoderm induction in early
Xenopus embryos. Cell, 79, 791-803.
Heinimann, K , MulLhaupt, B., Weber, W., Attenhofer, M., Scott, R. J., Fried, M.,
Martinoli. S., Muller, H. and Dobbie, Z. (1998) Phenotypic differences in familial
adenomatous polyposis based on APC gene mutation status. Gut, 43, 675-9.
Heise, C., Sampson-Johannes, A., Williams, A., McCormick, F., Von Hoff, D. D. and
Kirn, D. H. (1997) ONYX-015, an ElB gene-attenuated adenovirus, causes
267 tumor-specific cytolysis and antitumoral efficacy that can be augmented by
standard chemotherapeutic agents [see comments], Nat Med, 3, 639-45.
Hengge, U. R., Walker, P. S. and Vogel, J. C. (1996) Expression of naked DNA in
human, pig, and mouse skin, J C/m Invest, 97, 2911-6.
Hesdorffer, C., Ayello, J., Ward, M., Kaubisch, A., Vahdat, L., Balmaceda, C., Garrett,
T., Fetell, M., Reiss, R., Bank, A. and Antman, K. (1998) Phase I trial of
retro viral-mediated transfer of the human MDRl gene as marrow
chemoprotection in patients undergoing high-dose chemotherapy and autologous
stem-cell transplantation, J C/m Oncol, 16, 165-72.
Hirano, S., Nose, A., Hatta, K., Kawakami, A. and Takeichi, M. (1987) Calcium-
dependent cell-cell adhesion molecules (cadherins): subclass specificities and
possible involvement of actin bundles, J Cg// Biol, 105, 2501-10.
Horii, A., Nakatsuru, S., Miyoshi, Y., Ichii, S., Nagase, H., Ando, H., Yanagisawa, A.,
Tsuchiya, E., Kato, Y. and Nakamura, Y. (1992a) Frequent somatic mutations of
the APC gene in human pancreatic cancer. Cancer Res, 52, 6696-8.
Horii, A., Nakatsuru, S., Miyoshi, Y., Ichii, S., Nagase, H., Kato, Y., Yanagisawa, A. and
Nakamura, Y. (1992b) The APC gene, responsible for familial adenomatous
polyposis, is mutated in human gastric cancer. Cancer Res, 52, 3231-3.
Horsell, K. W., Merten, S., Clingan, P., King, D. W. and Morris, D. L. (1999)
Peritoneotomy and intraperitoneal chemotherapy in appendiceal and colorectal
cancer, AustNZJSurg, 69, 729-32.
Hoschuetzky, H., Aberle, H. and Kemler, R. (1994) Beta-catenin mediates the interaction
of the cadherin-catenin complex with epidermal growth factor receptor, J Cell
Biol, 127, 1375-80.
Hsiao, M., Tse, V., Carmel, J., Tsai, Y., Feigner, P. L., Haas, M. and Silverberg, G. D.
(1997) Intracavitary liposome-mediated p53 gene transfer into glioblastoma with
268 endogenous wild-type p53 in vivo results in tumor suppression and long-term
survival, Biochem Biophys Res Commun, 233, 359-64.
Huang, J., Papadopoulos, N., McKinley, A. J., Farrington, S. M., Curtis, L. J., Wyllie, A.
H., Zheng, S., Willson, J. K., Markowitz, S. D., Morin, P., Kinzler, K. W.,
Vogelstein, B. and Dunlop, M. G. (1996) APC mutations in colorectal tumors
with mismatch repair deficiency, Proc Natl Acad Sci USA, 93, 9049-54.
Huber, O., Korn, R., McLaughlin, J., Ohsugi, M., Herrmann, B. G. and Kemler, R. (1996)
Nuclear localization of beta-catenin by interaction with transcription factor LEF-1,
Mech Dev, 59, 3-10.
Hugh, T. J., Dillon, S. A., Taylor, B. A., Pignatelli, M., Poston, G. J. and Kinsella, A. R.
(1999) Cadherin-catenin expression in primary colorectal cancer: a survival
analysis, Br J Cancer, 80, 1046-51.
Hull, M. A., Booth, J. K., Tisbury, A., Scott, N., Bonifer, C., Markham, A. F. and Coletta,
P. L. (1999) Cyelooxygenase 2 is up-regulated and localized to macrophages in
the intestine of Min mice, Br J Cancer, 79, 1399-405.
Hulsken, J., Birchmeier, W. and Behrens, J. (1994) E-cadherin and APC compete for the
interaction with beta-catenin and the cyto skeleton, J Ce// Biol, 127,2061-9.
Huong, T. M., Harashima, H. and Kiwada, H. (1998) Complement dependent and
independent liposome uptake by peritoneal macrophages: cholesterol content
dependency, Biol Pharm Bull, 21, 969-73.
Ichii, S., Takeda, S., Horii, A., Nakatsuru, S., Miyoshi, Y., Emi, M., Fujiwara, Y.,
Koyama, K., Furuyama, J., Utsunomiya, J. and et al. (1993) Detailed analysis of
genetic alterations in colorectal tumors from patients with and without familial
adenomatous polyposis (FAP), Oncogene, 8, 2399-405.
Ilan, Y., Prakash, R., Davidson, A., Jona, Droguett, G., Horwitz, M. S., Chowdhury, N.
R. and Chowdhury, J. R. (1997) Oral tolerization to adenoviral antigens permits
269 long-term gene expression using recombinant adenoviral vectors, J Clin Invest,
99, 1098-106.
Jacoby, R. F., Marshall, D. J., Newton, M. A., Novakovic, K., Tutsch, K., Cole, C. E.,
Lubet, R A., Kelloff, G. J., Verma, A., Moser, A. R. and Dove, W. F. (1996)
Chemoprevention of spontaneous intestinal adenomas in the Ape Min mouse
model by the nonsteroidal anti-inflammatory drug piroxicam. Cancer Res, 56,
710-4.
Jais, P., Sabourin, J. C., Bombled, J., Rougier, P., Lasser, P., Duvillard, P., Benard, J. and
Bressac-de Paillerets, B. (1998) Absence of somatic alterations of the EBl gene
adenomatous polyposis coli-associated protein in human sporadic colorectal
cancers, Br J Cancer, 78, 1356-60.
Johnson, D. A., Gautsch, J. W., Sportsman, J. R. and Elder, J. H. (1984) Improved
technique utilizing nonfat dry milk for analysis of proteins and nucleic acids
transferred to nitrocellulose.. Gene Anal Tech., 1, 3.
Joslyn, G., Richardson, D. S., White, R. and Alber, T. (1993) Dimer formation by an N-
terminal coiled coil in the APC protein, Proc Natl Acad Sci USA, 90, 11109-13.
Jou, T. S., Stewart, D. B., Stappert, J., Nelson, W. J. and Marrs, J. A. (1995) Genetic and
biochemical dissection of protein linkages in the cadherin- catenin complex, Proc
Natl Acad Sci USA, 92, 5067-71.
Juwana, J. P., Henderikx, P., Mischo, A., Wadle, A., Fadle, N., Gerlach, K., Arends, J.
W., Hoogenboom, H., Pfreundschuh, M. and Renner, C. (1999) EB/RP gene
family encodes tubulin binding proteins, Int J Cancer, 81, 275-84.
Kagami, H., Atkinson, J. C., Michalek, S. M., Handelman, B., Yu, S., Baum, B. J. and
O'Connell, B. (1998) Repetitive adenovirus administration to the parotid gland:
role of immunological barriers and induction of oral tolerance. Hum Gene Ther, 9,
305-13.
270 Kanai, Y., Ochiai, A., Shibata, T., Oyama, T., Ushijima, S., Akimoto, S. and Hirohashi,
S. (1995) c-erbB-2 gene product directly associates with beta-catenin and
plakoglobin, Biochem Biophys Res Commun, 208, 1067-72.
Keighley, M. R., Grobler, S. and Bain, I. (1993) An audit of restorative proctocolectomy.
Gut, 34, 680-4.
Kikuchi, A., Aoki, Y., Sugaya, S., Serikawa, T., Takakuwa, K., Tanaka, K., Suzuki, N.
and Kikuchi, H. (1999) Development of novel cationic liposomes for efficient
gene transfer into peritoneal disseminated tumor. Hum Gene Ther, 10, 947-55.
Kinzler, K. W., Nilbert, M. C., Su, L. K., Vogelstein, B., Bryan, T. M., Levy, D. B.,
Smith, K. J., Preisinger, A. C., Hedge, P., McKechnie, D. and et al. (1991)
Identification of FAP locus genes from chromosome 5q21, Science, 253, 661-5.
Kinzler, K. W. and Vogelstein, B. (1996) Lessons from hereditary colorectal cancer. Cell,
87, 159-70.
Kitaeva, M. N., Grogan, L., Williams, J. P., Dimond, E., Nakahara, K., Hausner, P.,
DeNobile, J. W., Soballe, P. W. and Kirsch, I. R. (1997) Mutations in beta-catenin
are uncommon in colorectal cancer occurring in occasional replication error-
positive tumors. Cancer Res, 57, 4478-81.
Kleppe, K., Ohtsuka, E., Kleppe, R., Molineux, I. and Khorana, H. G. (1971) Studies on
polynucleotides. XCVI. Repair replications of short synthetic DNA's as catalyzed
by DNA polymerases, J Mol Biol, 56, 341-61.
Knudsen, K. A., Soler, A. P., Johnson, K. R. and Wheelock, M. J. (1995) Interaction of
alpha-actinin with the cadherin/catenin cell-cell adhesion complex via alpha-
catenin, J Biol, 130, 67-77.
Konishi, F., Maeda, H., Yamanishi, Y., Hiyama, K., Ishioka, S. and Yamakido, M. (1999)
Transcriptionally targeted in vivo gene therapy for carcinoembrionic antigen-
producing adenocarcinoma, Hiroshima J Med Sci, 48, 79-89.
271 Korinek, V., Barker, N., Morin, P. J., van Wichen, D., de Weger, R., Kinzler, K. W.,
Vogelstein, B. and Clevers, H. (1997) Constitutive transcriptional activation by a
beta-catenin-Tcf complex in APC-/- colon carcinoma [see comments]. Science,
275, 1784-7.
Kremer, E. J. and Perricaudet, M. (1995) Adenovirus and adeno-associated virus
mediated gene transfer, Br Med Bull, 51, 31-44.
Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of
bacteriophage T4, Nature, 227, 680-5.
Laken, S. J., Papadopoulos, N., Petersen, G. M., Gruber, S. B., Hamilton, S. R.,
Giardiello, F. M., Brensinger, J. D., Vogelstein, B. and Kinzler, K. W. (1999)
Analysis of masked mutations in familial adenomatous polyposis, Proc Natl Acad
Sci USA, 96, 2322-6.
Laken, S. J., Petersen, G. M., Gruber, S. B., Oddoux, C., Ostrer, H., Giardiello, F. M.,
Hamilton, S. R., Hampel, H., Markowitz, A., Klimstra, D., Jhanwar, S., Winawer,
S., Offit, K., Luce, M. C., Kinzler, K. W. and Vogelstein, B. (1997) Familial
colorectal cancer in Ashkenazim due to a hypermutable tract in APC, Nat Genet,
17, 79-83.
Lamlum, H., Ilyas, M., Rowan, A., Clark, S., Johnson, V., Bell, J., Frayling, L,
Efstathiou, J., Pack, K., Payne, S., Roylance, R., Gorman, P., Sheer, D., Neale, K.,
Phillips, R., Talbot, I., Bodmer, W. and Tomlinson, I. (1999) The type of somatic
mutation at APC in familial adenomatous polyposis is determined by the site of
the germline mutation: a new facet to Knudson's 'two-hit' hypothesis. Nature
Medicine, 5, 1071-1075.
Lehrman, S. (1999) Virus treatment questioned after gene therapy death [news]. Nature,
401,517-8.
272 Leister, L, Weith, A., Bruderlein, S., Cziepluch, C., Kangwanpong, D., Schlag, P. and
Schwab, M. (1990) Human colorectal cancer: high frequency of deletions at
chromosome lp35. Cancer Res, 50, 7232-5.
Levy, D. B., Smith, K. J., Beazer-Barclay, Y., Hamilton, S. R., Vogelstein, B. and
Kinzler, K. W. (1994) Inactivation of both APC alleles in human and mouse
tumors. Cancer Res, 54, 5953-8.
Lew, J. I., Vargish, R., Kim, R. K., Guo, Y., Michelassi, F. and Arenas, R. B. (2000)
Tumor Suppressor Gene Therapy And COX 2 Inhibition Additively Reduces
Intestinal Neoplasia., Gastroenterology, 118, Abstract 2458.
Li, S., Tseng, W. C., Stolz, D. B., Wu, S. P., Watkins, S. C. and Huang, L. (1999)
Dynamic changes in the characteristics of cationic lipidic vectors after exposure to
mouse serum: implications for intravenous lipofection. Gene Ther, 6, 585-94.
Licht, T., Gottesman, M. M. and Pastan, I. (1995) Transfer of the MDRl (multidrug
resistance) gene: protection of hematopoietic cells from cytotoxic chemotherapy,
and selection of transduced cells in vivo. Cytokines Mol Ther, 1, 11-20.
Liu, B., Parsons, R. E., Hamilton, S. R., Petersen, G. M., Lynch, H. T., Watson, P.,
Markowitz, S., Willson, J, K , Green, J., de la Chapelle, A. and et al. (1994)
hMSH2 mutations in hereditary nonpolyposis colorectal cancer kindreds. Cancer
Res, 54,4590-4.
Liu, T. J., el-Naggar, A. K., McDonnell, T. J., Steck, K. D., Wang, M., Taylor, D. L. and
dayman, G. L. (1995a) Apoptosis induction mediated by wild-type p53
adenoviral gene transfer in squamous cell carcinoma of the head and neck. Cancer
Res, 55,3117-22.
Liu, Y., Liggitt, D., Zhong, W., Tu, G., Gaensler, K. and Debs, R. (1995b) Cationic
liposome-mediated intravenous gene delivery, J Biol Chem, 270,24864-70.
273 Lochmuller, H., Jani, A., Huard, J., Prescott, S., Simoneau, M., Massie, B., Karpati, G.
and Acsadi, G. (1994) Emergence of early region 1-containing replication-
competent adenovirus in stocks of replication-defective adenovirus recombinants
(delta El + delta E3) during multiple passages in 293 cells. Hum Gene Ther, 5,
1485-91.
Luongo, C., Moser, A. R., Gledhill, S. and Dove, W. F, (1994) Loss of Apc+ in intestinal
adenomas from Min mice. Cancer Res, 54, 5947-52.
MacPhee, M., Chepenik, K. P., Liddell, R. A., Nelson, K. K., Siracusa, L. D. and
Buchberg, A. M. (1995) The secretory phospho lipase A2 gene is a candidate for
the Moml locus, a major modifier of ApcMin-induced intestinal neoplasia. Cell,
81, 957-66.
Mahmoud, N. N., Bilinski, R. T., Churchill, M. R., Edelmann, W., Kucherlapati, R. and
Bertagnolli, M. M. (1999a) Genotype-phenotype correlation in murine Ape
mutation: differences in enterocyte migration and response to sulindac. Cancer
Res, 59, 353-9.
Mahmoud, N. N., Boolbol, S. K., Bilinski, R. T., Martucci, C., Chadburn, A. and
Bertagnolli, M. M. (1997) Ape gene mutation is associated with a dominant-
negative effect upon intestinal cell migration. Cancer Res, 57, 5045-50.
Mahmoud, N. N., Boolbol, S. K., Dannenberg, A. J., Mestre, J. R., Bilinski, R. T.,
Martucci, C., Newmark, H. L., Chadburn, A. and Bertagnolli, M. M. (1998a) The
sulfide metabolite of sulindac prevents tumors and restores enterocyte apoptosis in
a murine model of familial adenomatous polyposis. Carcinogenesis, 19, 87-91.
Mahmoud, N. N., Dannenberg, A. J., Bilinski, R. T., Mestre, J. R, Chadburn, A.,
Churchill, M., Martucci, C. and Bertagnolli, M. M. (1999b) Administration of an
unconjugated bile acid increases duodenal tumors in a murine model of familial
adenomatous polyposis. Carcinogenesis, 20, 299-303.
274 Mahmoud, N. N., Dannenberg, A. J., Mestre, J., Bilinski, R. T., Churchill, M. R.,
Martucci, G., Newmark, H. and Bertagnolli, M. M. (1998b) Aspirin prevents
tumors in a murine model of familial adenomatous polyposis. Surgery, 124, 225-
31.
Mandl, M., Paffenholz, R., Friedl, W., Caspari, R., Sengteller, M. and Propping, P. (1994)
Frequency of common and novel inactivating APC mutations in 202 families v^ith
familial adenomatous polyposis. Hum Mol Genet, 3, 181-4.
Mann, B., Gelos, M., Siedow, A., Hanski, M. L., Gratchev, A., Ilyas, M., Bodmer, W. F.,
Moyer, M. P., Riecken, E. O., Buhr, H. J. and Hanski, C. (1999) Target genes of
beta-catenin-T cell-factor/lymphoid-enhancer-factor signaling in human colorectal
carcinomas, Proc Natl Acad Soi USA, 96, 1603-8.
Markowitz, D., Goff, S. and Bank, A. (1988) Construction and use of a safe and efficient
amphotropic packaging cell line. Virology, 167, 400-6.
Matsumine, A., Ogai, A., Senda, T., Okumura, N., Satoh, K., Baeg, G. H., Kawahara, T.,
Kobayashi, S., Okada, M., Toyoshima, K. and Akiyama, T. (1996) Binding of
APC to the human homolog of the Drosophila discs large tumor suppressor
protein [see comments]. Science, 272, 1020-3.
McCrea, P. D., Turck, C. W. and Gumbiner, B. (1991) A homo log of the armadillo
protein in Drosophila (plakoglobin) associated with E-cadherin, Science, 254,
1359-61.
Mercola, K. E., Stang, H. D., Browne, J., Salser, W. and Cline, M. J. (1980) Insertion of a
new gene of viral origin into bone marrow cells of mice. Science, 208, 1033-5.
Mitani, K., Wakamiya, M. and Caskey, C. T. (1993) Long-term expression of retroviral-
transduced adenosine deaminase in human primitive hematopoietic progenitors.
Hum Gene Ther, 4, 9-16.
275 Miyaki, M., Konishi, M., Kikuchi-Yanoshita, R., Enomoto, M., Tanaka, K., Takahashi,
H., Muraoka, M., Mori, T., Konishi, F. and Iwama, T. (1993) Coexistence of
somatic and germ-line mutations of APC gene in desmoid tumors from patients
with familial adenomatous polyposis. Cancer Res, 53, 5079-82.
Miyoshi, Y., Ando, H., Nagase, H., Nishisho, L, Horii, A., Miki, Y., Mori, T.,
Utsunomiya, J., Baba, S., Petersen, G. and et al. (1992a) Germ-line mutations of
the APC gene in 53 famihal adenomatous polyposis patients, Proc Natl Acad Sci
USA, 89, 4452-6.
Miyoshi, Y., Nagase, H., Ando, H., Horii, A., Ichii, S., Nakatsuru, S., Aoki, T., Miki, Y.,
Mori, T. and Nakamura, Y. (1992b) Somatic mutations of the APC gene in
colorectal tumors: mutation cluster region in the APC gene. Hum Mol Genet, 1,
229-33.
Molenaar, M., van de Wetering, M., Oosterwegel, M., Peterson-Maduro, J., Godsave, S.,
Korinek, V., Roose, J., Destree, O. and Clevers, H. (1996) XTcf-3 transcription
factor mediates beta-catenin-induced axis formation in Xenopus embryos. Cell,
86, 391-9.
Monakov, O., Shemesh, E., Bar-Meir, S. and Gazit, E. (1998) Novel nonsense mutation
in exon 15 of the APC gene in one Jewish family. Hum Mutat, Suppl, S55.
Morgan, R. J., Jr., Braly, P., Cecchi, G., Leong, L., Shibata, S., Margolin, K., Somlo, G.,
McNamara, M., Longmate, J., Schinke, S., Raschko, J., Nagasawa, S., Kogut, N.,
Parker, P., Stein, A., Cho, J., Smith, E., Coluzzi, P., Najera, L., Johnson, D.,
Womack, E. and Doroshow, J. H. (1999) Phase II trial of intraperitoneal cisplatin
with intravenous doxorubicin and cyclophosphamide in previously untreated
patients with advanced ovarian cancer-long-term follow-up, Gynecol Oncol, 75,
419-26.
276 Morin, P. J., Sparks, A. B., Korinek, V., Barker, N., Clevers, H., Vogelstein, B. and
Kinzler, K. W. (1997) Activation of beta-catenin-Tcf signaling in colon cancer by
mutations in beta-catenin or APC [see comments]. Science, 275, 1787-90.
Morin, P. J., Vogelstein, B. and Kinzler, K. W. (1996) Apoptosis and APC in colorectal
tumorigenesis, Proc Natl Acad Sci USA, 93, 7950-4.
Moscow, J. A., Huang, H., Carter, C., Hines, K., Zujewski, J., Cusack, G., Chow, C.,
Venzon, D., Sorrentino, B., Chiang, Y., Goldspiel, B., Leitman, S., Read, E. J.,
Abati, A., Gottesman, M. M., Pastan, I., Sellers, S., Dunbar, C. and Cowan, K. H.
(1999) Engraftment of MDRl and NeoR gene-transduced hematopoietic cells
after breast cancer chemotherapy. Blood, 94, 52-61.
Moser, A. R., Pitot, H. C. and Dove, W. F. (1990) A dominant mutation that predisposes
to multiple intestinal neoplasia in the mouse. Science, 247, 322-4.
Mujoo, K., Maneval, D. C., Anderson, S. C. and Gutterman, J. U. (1996) Adenoviral-
mediated p53 tumor suppressor gene therapy of human ovarian carcinoma.
Oncogene, 12, 1617-23.
Muller, T., Choidas, A., Reichmann, E. and Ullrich, A. (1999) Phosphorylation and fi’ee
pool of beta-catenin are regulated by tyrosine kinases and tyrosine phosphatases
during epithelial cell migration, C/zew, 274, 10173-83.
Mulligan, R. C., Howard, B. H. and Berg, P. (1979) Synthesis of rabbit beta-globin in
cultured monkey kidney cells following infection with a SV40 beta-globin
recombinant genome. Nature, 277,108-14.
Mullis, K. B. and Faloona, F. A. (1987) Specific synthesis of DNA in vitro via a
polymerase-catalyzed chain reaction. Methods Enzymol, 155, 335-50.
Munemitsu, S., Albert, I., Souza, B., Rubinfeld, B. and Polakis, P. (1995) Regulation of
intracellular beta-catenin levels by the adenomatous polyposis coli (APC) tumor-
suppressor protein, Proc Natl Acad Sci USA, 92, 3046-50.
277 Munemitsu, S., Souza, B., Muller, O., Albert, L, Rubinfeld, B. and Polakis, P. (1994) The
APC gene product associates with microtubules in vivo and promotes their
assembly in vitro. Cancer Res, 54, 3676-81.
Nabel, G. J., Nabel, E. G., Yang, Z. Y., Fox, B. A., Plautz, G. E., Gao, X., Huang, L.,
Shu, S., Gordon, D. and Chang, A. E. (1993) Direct gene transfer with DNA-
liposome complexes in melanoma: expression, biologic activity, and lack of
toxicity in humans, Proc Natl Acad Sci USA, 90, 11307-11.
Nagafuchi, A. and Takeichi, M. (1988) Cell binding fimction of E-cadherin is regulated
by the cytoplasmic domain, Embo J, 7, 3679-84.
Nagase, H. and Nakamura, Y. (1993) Mutations of the APC (adenomatous polyposis coli)
gene. Hum Mutat, 2,425-34.
Naldini, L., Blomer, U., Gallay, P., Ory, D., Mulligan, R., Gage, F. H., Verma, I. M. and
Trono, D. (1996) In vivo gene delivery and stable transduction of nondividing
cells by a lentivLral vector [see comments]. Science, 272, 263-7.
Nathke, I. S., Adams, C. L., Polakis, P., Sellin, J. H. and Nelson, W. J. (1996) The
adenomatous polyposis coli tumor suppressor protein localizes to plasma
membrane sites involved in active cell migration, J Ce/Z Biol, 134, 165-79.
Nguyen, D. M., Spitz, F. R., Yen, N., Cristiano, R. J. and Roth, J. A. (1996) Gene therapy
for lung cancer: enhancement of tumor suppression by a combination of
sequential systemic cisplatin and adenovirus-mediated p53 gene transfer, J Thorac
Cardiovasc Surg, 112, 1372-6; discussion 1376-7.
Niedzinski, E. J., Bennett, M. J., Olson, D. C. and Nantz, M. H. (2000) Gastroprotection
of DNA with a synthetic cholic acid analog [In Process Citation], Lipids, 35, 721-
7.
278 Nishisho, L, Nakamura, Y., Miyoshi, Y., Miki, Y., Ando, H., Horii, A., Koyama, K.,
Utsunomiya, J., Baba, S. and Hedge, P. (1991) Mutations of chromosome 5q21
genes in FAP and colorectal cancer patients. Science, 253, 665-9.
Northover, J. M. (1987) Activities of the Leeds Castle Polyposis Group, Semin Surg
Oncol, 3, 118-9.
Nose, A., Tsuji, K. and Takeichi, M. (1990) Localization of specificity determining sites
in cadherin cell adhesion molecules. Cell, 61, 147-55.
Novelli, M. R., Wasan, H., Rosewell, L, Bee, J., Tomlinson, I. P., Wright, N. A. and
Bodmer, W. F. (1999) Tumor burden and clonality in multiple intestinal neoplasia
mouse/normal mouse aggregation chimeras, Proc Natl Acad Sci USA, 96, 12553-
8.
Nugent, K. P., Farmer, K. C., Spigelman, A. D., Williams, C. B. and Phillips, R. K.
(1993a) Randomized controlled trial of the effect of sulindac on duodenal and
rectal polyposis and cell proliferation in patients with familial adenomatous
polyposis, Br J Surg, 80, 1618-9.
Nugent, K. P. and Phillips, R. K. (1992) Rectal cancer risk in older patients with familial
adenomatous polyposis and an ileorectal anastomosis: a cause for concern, Br J
Surg, 79, 1204-6.
Nugent, K. P., Spigelman, A. D. and Phillips, R. K. (1993b) Life expeetaney after
colectomy and ileorectal anastomosis for familial adenomatous polyposis. Dis
Colon Rectum, 36, 1059-62.
Nuyttens, J. J., Rust, P. F., Thomas, C. R., Jr. and Turrisi, A. T., 3rd (2000) Surgery
versus radiation therapy for patients with aggressive fibromatosis or desmoid
tumors: A comparative review of 22 articles. Cancer, 88, 1517-23.
Ogawa, N., Fujiwara, T., Kagawa, S., Nishizaki, M., Morimoto, Y., Tanida, T., Hizuta,
A., Yasuda, T., Roth, J. A. and Tanaka, N. (1997) Novel combination therapy for
279 human colon cancer with adenovirus- mediated wild-type p53 gene transfer and
DNA-damaging chemotherapeutic agent, Int J Cancer, 73, 367-70.
Olschwang, S., Laurent-Puig, P., Groden, J., White, R. and Thomas, G. (1993a) Germ-
line mutations in the first 14 exons of the adenomatous polyposis coli (APC) gene.
Am J Hum Genet, 52, 273-9.
Olschwang, S., Tiret, A., Laurent-Puig, P., Muleris, M., Parc, R. and Thomas, G. (1993b)
Restriction of ocular fundus lesions to a specific subgroup of APC mutations in
adenomatous polyposis coli patients. Cell, 75, 959-68.
Omstein, L. (1964) Disc electrophoresis-1. Background and theory, Ann. N. Y. Acad. Sci,
121, 321-.
Orsulic, S. and Peifer, M. (1996) An in vivo structure-function study of armadillo, the
beta-catenin homologue, reveals both separate and overlapping regions of the
protein required for cell adhesion and for wingless signaling, J Cell Biol, 134,
1283-300.
Oshima, M., Dinchuk, J. E., Kargman, S. L., Oshima, H., Hancock, B., Kwong, E.,
Trzaskos, J. M., Evans, J. F. and Taketo, M. M. (1996) Suppression of intestinal
polyposis in Ape delta? 16 knockout mice by inhibition of cyclooxygenase 2
(COX-2), Cell, 87, 803-9.
Oshima, M., Oshima, H., Kitagawa, K., Kobayashi, M., Itakura, C. and Taketo, M. (1995)
Loss of Ape heterozygosity and abnormal tissue building in nascent intestinal
polyps in mice carrying a truncated Ape gene, Proc Natl Acad Sci USA, 92,
4482-6.
O’Sullivan, M. J., Mulcahy, T. M., Cambell, J., O’Suilleabhain, C. B., Kirwan, W. O.,
Doyle, C. T. and McCarthy, T. V. (1998) Detection of five novel germline
mutations of the APC gene in Irish familial adenomatous polyposis families. Hum
Mutat, Suppl, S251-3.
280 Palmirotta, R., Curia, M. C., Esposito, D. L., Valanzano, R., Messerini, L., Ficari, F.,
Brandi, M. L., Tonelli, F., Mariani-Costantini, R., Battista, P. and et al (1995)
Novel mutations and inactivation of both alleles of the APC gene in desmoid
tumors. Hum Mol Genet, 4, 1979-81.
Pastan, I. and Gottesman, M. M. (1991) Multidrug resistance, Annu Rev Med, 42, 277-86.
Patel, S. R., Evans, H. L. and Benjamin, R. S. (1993) Combination chemotherapy in adult
desmoid tumors. Cancer, 72, 3244-7.
Peifer, M., Berg, S. and Reynolds, A. B. (1994) A repeating amino acid motif shared by
proteins with diverse cellular roles [letter]. Cell, 76, 789-91.
Philips, R., Spigelman, A. and Thompson, J. (1994) Familial Adenomatous Polyposis and
other polyposis syndromes, Edward Arnold, London.
Piazza, G. A., Alberts, D. S., Hixson, L. J., Paranka, N. S., Li, H., Finn, T., Bogert, C.,
Guillen, J. M., Brendel, K., Gross, P. H., Sperl, G., Ritchie, J., Burt, R. W.,
Ellsworth, L., Ahnen, D. J. and Pamukcu, R. (1997a) Sulindac sulfone inhibits
azoxymethane-induced colon carcinogenesis in rats without reducing
prostaglandin levels. Cancer Res, 57, 2909-15.
Piazza, G. A., Rahm, A. K., Firm, T. S., Fryer, B. H., Li, H., Stoumen, A. L., Pamukcu, R.
and Ahnen, D. J. (1997b) Apoptosis primarily accounts for the growth-inhibitory
properties of sulindac metabolites and involves a mechanism that is independent
of cyclooxygenase inhibition, cell cycle arrest, and p53 induction. Cancer Res, 57,
2452-9.
Plail, R. O., Bussey, H. J., Glazer, G. and Thomson, J. P. (1987) Adenomatous polyposis:
an association with carcinoma of the thyroid, B r J Surg, 74, 377-80.
Polakis, P. (1995) Mutations in the APC gene and their implications for protein structure
and function, Curr Opin Genet Dev, 5, 66-71.
281 Powell, S. M., Zilz, N., Beazer-Barclay, Y., Bryan, T. M., Hamilton, S. R., Thibodeau, S.
N., Vogelstein, B. and Kinzler, K. W. (1992) APC mutations occur early during
colorectal tumorigenesis. Nature, 359, 235-7.
Preobrazhensky, A. A. (1993) Electrophoresis of neurochordins, a family of high
molecular weight neural tissue glycoproteins, on horizontal submerged agarose
gels in the presence of dodecyl sulfate. Anal Biochem, 209, 315-7.
Quelle, D. E., Ashmun, R. A., Shurtleff, S. A., Kato, J. Y., Bar-Sagi, D., Roussel, M. F.
and Sherr, C. J. (1993) Overexpression of mouse D-type cyclins accelerates G1
phase in rodent fibroblasts. Genes Dev, 7, 1559-71.
Rabinowitz, J. E. and Samulski, J. (1998) Adeno-associated virus expression systems for
gene transfer, Curr Opin Biotechnol, 9, 470-5.
Ram, Z., Culver, K. W., Oshiro, E. M., Viola, J. J., DeVroom, H. L., Otto, E., Long, Z.,
Chiang, Y., McGarrity, G. J., Muul, L. M., Katz, D., Blaese, R. M. and Oldfield,
E. H. (1997) Therapy of malignant brain tumors by intratumoral implantation of
retroviral vector-producing cells [see comments], Nat Med, 3, 1354-61.
Ram, Z., Culver, K. W., Walbridge, S., Blaese, R. M. and Oldfield, E. H. (1993) In situ
retroviral-mediated gene transfer for the treatment of brain tumors in rats [see
comments]. Cancer Res, 53, 83-8.
Ram, Z., Walbridge, S., Shawker, T., Culver, K. W., Blaese, R. M. and Oldfield, E. H.
(1994) The effect of thymidine kinase transduction and ganciclovir therapy on
tumor vasculature and growth of 9L gliomas in rats, JNeurosurg, 81, 256-60.
Reddy, B. S., Rao, C. V. and Seibert, K. (1996) Evaluation of cyclooxygenase-2 inhibitor
for potential chemopreventive properties in colon carcinogenesis. Cancer Res, 56,
4566-9.
282 Regillo, C. D., Eagle, R. C., Jr., Shields, J. A., Shields, C. L. and Arbizo, V. V. (1993)
Histopathologic findings in congenital grouped pigmentation of the retina,
Ophthalmology, 100,400-5.
Reitmair, A. H., Cai, J. C., Bjerknes, M., Redston, M., Cheng, H., Find, M. T., Hay, K.,
Mitri, A., Bapat, B. V., Mak, T. W. and Gallinger, S. (1996a) MSH2 deficiency
contributes to accelerated AFC-mediated intestinal tumorigenesis. Cancer Res, 56,
2922-6.
Reitmair, A. H., Redston, M., Cai, J. C., Chuang, T. C., Bjerknes, M., Cheng, H., Hay,
K., Gallinger, S., Bapat, B. and Mak, T. W. (1996b) Spontaneous intestinal
carcinomas and skin neoplasms in Msh2-deficient mice. Cancer Res, 56, 3842-9.
Riggleman, B., Wieschaus, E. and Schedl, F. (1989) Molecular analysis of the armadillo
locus: uniformly distributed transcripts and a protein with novel internal repeats
are associated with a Drosophila segment polarity gene. Genes Dev, 3, 96-113.
Rizzuto, G., Cappelletti, M., Maione, D., Savino, R., Lazzaro, D., Costa, F., Mathiesen, I.,
Cortese, R., Ciliberto, G., Laufer, R., La Monica, N. and Fattori, E. (1999)
Efficient and regulated erythropoietin production by naked DNA injection and
muscle electroporation, Proc Natl Acad Sci USA, 96, 6417-22.
Roth, J. A., Nguyen, D., Lawrence, D. D., Kemp, B. L., Carrasco, C. H., Person, D. Z.,
Hong, W. K., Komaki, R., Lee, J. J., Nesbitt, J. C., Fisters, K. M., Futnam, J. B.,
Schea, R., Shin, D. M., Walsh, G. L., Dolormente, M. M., Han, C. L, Martin, F.
D., Yen, N., Xu, K., Stephens, L. C., McDonnell, T. J., Mukhopadhyay, T. and
Cai, D. (1996) Retrovirus-mediated wild-type p53 gene transfer to tumors of
patients with lung cancer [see comments], Nat Med, 2, 985-91.
Rozen, F., Shomrat, R., Strul, H., Naiman, T., Karminsky, N., Legum, C. and Orr-
Urtreger, A. (1999) Frevalence of the I1307K AFC gene variant in Israeli Jews of
283 differing ethnic origin and risk for colorectal cancer [see comments],
Gastroenterology, 116, 54-7.
Rubin, J., Galanis, E., Pitot, H. C., Richardson, R. L., Burch, P. A., Charboneau, J. W.,
Reading, C. C., Lewis, B. D., Stahl, S., Akporiaye, E. T. and Harris, D. T. (1997)
Phase I study of immunotherapy of hepatic métastasés of colorectal carcinoma by
direct gene transfer of an allogeneic histocompatibility antigen, HLA-B7, Gene
Ther, 4,419-25.
Rubinfeld, B., Albert, I., Porfiri, E., Fiol, C., Munemitsu, S. and Polakis, P. (1996)
Binding of GSK3beta to the APC-beta-catenin complex and regulation of complex
assembly [see comments]. Science, 272, 1023-6.
Russell, S. J. (1994) Replicating vectors for cancer therapy: a question of strategy, Semin
Cancer Biol, 5, 437-43.
Sacco, M. G., Barbieri, O., Piccini, D., Noviello, E., Zoppe, M., Zucchi, I., Frattini, A.,
Villa, A. and Vezzoni, P. (1998) In vitro and in vivo antisense-mediated growth
inhibition of a mammary adenocarcinoma from MMTV-neu transgenic mice,
Gene Ther, 5, 388-93.
Sagawa, H., Tazuma, S. and Kajiyama, G. (1993) Protection against hydrophobic bile
salt-induced cell membrane damage by liposomes and hydrophilic bile salts, Am J
Physiol, 264, G835-9.
Saiki, R. K., Scharf, S., Faloona, F., Mullis, K. B., Horn, G. T., Erlich, H. A. and
Arnheim, N. (1985) Enzymatic amplification of beta-globin genomic sequences
and restriction site analysis for diagnosis of sickle cell anemia, Science, 230,
1350-4.
Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989a) In Molecular Cloning: A Laboratory
ManualCold Spring Harbour Laboratory Press, New York, pp. 7.46-7.48.
284 Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989b) In Molecular Cloning: A
Laboratory Manual, Vol. 3 Cold Spring Harbor Laboratory, New York, pp. 18.67.
Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989c) In Molecular Cloning: A Laboratory
Manual, Vol. 3 Cold Spring Harbor Laboratory, New York, pp. 18.55.
Sammons, D. W., Adams, L. D. and Nishizawa, E. E. (1981) Ultrasensitive silver-based
color staining of polypeptides in polyacrylamide gels.. Electrophoresis, 2, 135-
141.
Samowitz, W. S., Powers, M. D., Spirio, L. N., Nollet, F., van Roy, F. and Slattery, M. L.
(1999) Beta-catenin mutations are more frequent in small colorectal adenomas
than in larger adenomas and invasive carcinomas. Cancer Res, 59, 1442-4.
Sandberg, J., Lau, C., Jacomino, M., Finegold, M. and Henning, S. (1994) Improving
access to intestinal stem cells as a step toward intestinal gene transfer. Human
Gene Therapy, 5, 323-329.
Sano, H., Kawahito, Y., Wilder, R. L., Hashiramoto, A., Mukai, S., Asai, K., Kimura, S.,
Kato, H., Kondo, M. and Hla, T. (1995) Expression of cyclooxygenase-1 and -2 in
human colorectal cancer. Cancer Res, 55, 3785-9.
Schagen, F. H., Rademaker, H. J., Fallaux, F. J. and Hoeben, R. C. (2000) Insertion
vectors for gene therapy [editorial]. Gene Ther, 7,271-2.
Schiedner, G., Morral, N., Parks, R. J., Wu, Y., Koopmans, S. C., Langston, C., Graham,
F. L., Beaudet, A. L. and Kochanek, S. (1998) Genomic DNA transfer with a
high-capacity adenovirus vector results in improved in vivo gene expression and
decreased toxicity [published erratum appears in Nat Genet 1998 Mar;18(3):298],
Nat Genet, 18, 180-3.
Schnitzler, M., Dwight, T. and Robinson, B. G. (1996) Sulindac increases the expression
of APC mRNA in malignant colonic epithelial cells: an in vitro study. Gut, 38,
707-13.
285 Schnitzler, M., Koorey, D., Dwight, T., Tomaras, C., Macrae, F., Marsh, D. and
Robinson, B. (1998) Frequency of codon 1061 and codon 1309 APC mutations in
Australian familial adenomatous polyposis patients. Hum Mutat, Suppl, S56-7.
Schwartz, B., Benoist, C., Abdallah, B., Rangara, R., Hassan, A., Scherman, D. and
Demeneix, B. A. (1996) Gene transfer by naked DNA into adult mouse brain.
Gene Ther, 3,405-11.
Shampay, J., Szostak, J. W. and Blackburn, E. H. (1984) DNA sequences of telomeres
maintained in yeast. Nature, 310, 154-7.
Shao, J., Fujiwara, T., Kadowaki, Y., Fukazawa, T., Waku, T., Itoshima, T., Yamatsuji,
T., Nishizaki, M., Roth, J. A. and Tanaka, N. (2000) Overexpression of the wild-
type p53 gene inhibits NF-kappaB activity and synergizes with aspirin to induce
apoptosis in human colon cancer cells. Oncogene, 19, 726-36.
Shibamoto, S., Hayakawa, M., Takeuchi, K., Hori, T., Oku, N., Miyazawa, K., Kitamura,
N., Takeichi, M. and Ito, F. (1994) Tyrosine phosphorylation of beta-catenin and
plakoglobin enhanced by hepatocyte growth factor and epidermal growth factor in
human carcinoma cells. Cell Adhes Commun, 1, 295-305.
Shibata, H., Toyama, K., Shioya, H., Ito, M., Hirota, M., Hasegawa, S., Matsumoto, H.,
Takano, H., Akiyama, T., Toyoshima, K., Kanamaru, R., Kanegae, Y., Saito, I.,
Nakamura, Y., Shiba, K. and Noda, T. (1997) Rapid colorectal adenoma
formation initiated by conditional targeting of the Ape gene. Science, 278, 120-3.
Shih, I. M., Yu, J., He, T. C., Vogelstein, B. and Kinzler, K. W. (2000) The beta-catenin
binding domain of adenomatous polyposis coli is sufficient for tumor suppression.
Cancer Res, 60, 1671-6.
Shoemaker, A. R., Luongo, C., Moser, A. R., Marton, L. J. and Dove, W. F. (1997)
Somatic mutational mechanisms involved in intestinal tumor formation in Min
mice. Cancer Res, 57, 1999-2006.
286 Shoemaker, A. R., Moser, A. R. and Dove, W. F. (1995) N-ethyl-N-nitrosourea treatment
of multiple intestinal neoplasia (Min) mice: age-related effects on the formation of
intestinal adenomas, cystic crypts, and epidermoid cysts. Cancer Res, 55, 4479-
85.
Siegfried, E., Wilder, E. L. and Perrimon, N. (1994) Components of wingless signalling
in Drosophila, Nature, 367, 76-80.
Skapek, S. X., Hawk, B. J., Hoffer, F. A., Dahl, G. V., Granowetter, L., Gebhardt, M. C.,
Ferguson, W. S. and Grier, H. E. (1998) Combination chemotherapy using
vinblastine and methotrexate for the treatment of progressive desmoid tumor in
children [see comments], J C/m Oncol, 16, 3021-7.
Smith, A. J., Stem, H. S., Fenner, M., Hay, K., Mitri, A., Bapat, B. V. and Gallinger, S.
(1994) Somatic APC and K-ras codon 12 mutations in aberrant crypt foci from
human colons. Cancer Res, 54, 5527-30.
Smith, K. J., Johnson, K. A., Bryan, T. M., Hill, D. E., Markowitz, S., Willson, J. K.,
Paraskeva, C., Petersen, G. M., Hamilton, S. R., Vogelstein, B. and et al. (1993)
The APC gene product in normal and tumor cells, Proc Natl Acad Sci USA, 90,
2846-50.
Smith, M. E., Marsh, S. G., Bodmer, J. G., Gelsthorpe, K. and Bodmer, W. F. (1989)
Loss of HLA-A,B,C allele products and lymphocyte fimction-associated antigen 3
in colorectal neoplasia, Proc Natl Acad Sci USA, 86, 5557-61.
Smith, P. K., Krohn, R. I., Hermanson, G. T., Mallia, A. K., Gartner, F. H., Provenzano,
M. D., Fujimoto, E. K., Goeke, N. M., Olson, B. J. and Klenk, D. C. (1985)
Measurement of protein using bicinchoninic acid [published erratum appears in
Anal Biochem 1987 May 15;163(1):279], vfW Biochem, 150, 76-85.
Smits, R., Kielman, M. F., Breukel, C., Zurcher, C., Neufeld, K., Jagmohan-Changur, S.,
Holland, N., van Dijk, J., White, R., Edelmann, W., Kucherlapati, R., Khan, P. M.
287 and Fodde, R. (1999) Apcl638T: a mouse model delineating critical domains of
the adenomatous polyposis coli protein involved in tumorigenesis and
development, Genes Dev, 13, 1309-21.
Smits, R., van der Houven van Oordt, W., Luz, A., Zurcher, C., Jagmohan-Changur, S.,
Breukel, C., Khan, P. M. and Fodde, R. (1998) Apcl638N: a mouse model for
familial adenomatous polyposis-associated desmoid tumors and cutaneous cysts.
Gastroenterology, 114, 275-83.
Sommers, C. L., Gelmann, E. P., Kemler, R., Co win, P. and Byers, S. W. (1994)
Alterations in beta-catenin phosphorylation and plakoglobin expression in human
breast cancer cells. Cancer Res, 54, 3544-52.
Soravia, C., Berk, T., McLeod, R. S. and Cohen, Z. (2000) Desmoid disease in patients
with familial adenomatous polyposis. Dis Colon Rectum, 43, 363-9.
Sparks, A. B., Morin, P. J., Vogelstein, B. and Kinzler, K. W. (1998) Mutational analysis
of the APC/beta-catenin/Tcf pathway in colorectal cancer. Cancer Res, 58, 1130-
4.
Spigelman, A. D., Williams, C. B., Talbot, I. C., Domizio, P. and Phillips, R. K. (1989)
Upper gastrointestinal cancer in patients with familial adenomatous polyposis.
Lancet, 2, 783-5.
Spirio, L., Olschwang, S., Groden, J., Robertson, M., Samowitz, W., Joslyn, G., Gelbert,
L., Thliveris, A., Carlson, M., Otterud, B. and et al. (1993) Alleles of the APC
gene: an attenuated form of familial polyposis. Cell, 75, 951-7.
Steele, R. J., Thompson, A. M., Hall, P. A. and Lane, D. P. (1998) The p53 tumour
suppressor gene, Br J Surg, 85, 1460-7.
Steinbach, G., Lynch, P. M., Phillips, R. K., Wallace, M. H., Hawk, E., Gordon, G. B.,
Wakabayashi, N., Saunders, B., Shen, Y., Fujimura, T., Su, L. K. and Levin, B.
288 (2000) The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial
adenomatous p:\ypos\s, N EnglJ Med, 342, 1946-52.
Stella, A., Resta, N., Gentile, M., Susca, F., Mareni, C., Montera, M. P. and Guanti, G.
(1993) Exclusion of the APC gene as the cause of a variant form of familial
adenomatous polyposis (FAP) [see comments]. Am J Hum Genet, 53, 1031-7.
Stribling, R., Brunette, E., Liggitt, D., Gaensler, K. and Debs, R. (1992) Aerosol gene
delivery in vivo, Proc Natl Acad Sci USA, 89,11277-81.
Su, L. K., Burrell, M., Hill, D. E., Gyuris, J., Brent, R., Wiltshire, R., Trent, J.,
Vogelstein, B. and Kinzler, K. W. (1995) APC binds to the novel protein EBl,
Cancer Res, 55, 2972-7.
Su, L. K., Johnson, K. A., Smith, K. J., Hill, D. E., Vogelstein, B. and Kinzler, K. W.
(1993a) Association between wild type and mutant APC gene products. Cancer
Res, 53, 2728-31.
Su, L. K., Kinzler, K. W., Vogelstein, B., Preisinger, A. C., Moser, A. R., Luongo, C.,
Gould, K. A. and Dove, W. F. (1992) Multiple intestinal neoplasia caused by a
mutation in the murine homolog of the APC gene [published erratum appears in
Science 1992 May 22;256(5060):1114], Science, 256, 668-70.
Su, L. K., Vogelstein, B. and Kinzler, K. W. (1993b) Association of the APC tumor
suppressor protein with catenins. Science, 262, 1734-7.
Sullenger, B. A. and Cech, T. R. (1994) Ribozyme-mediated repair of defective mRNA
by targeted, trans-splicing. Nature, 371, 619-22.
Swisher, S. G., Roth, J. A., Nemunaitis, J., Lawrence, D. D., Kemp, B. L., Carrasco, C.
H., Connors, D. G., El-Naggar, A. K., Fossella, F., Glisson, B. S., Hong, W. K.,
Khuri, F. R , Kurie, J. M., Lee, J. J., Lee, J. S., Mack, M., Merritt, J. A., Nguyen,
D. M., Nesbitt, J. C., Perez-Soler, R., Pisters, K. M., Putnam, J. B., Jr., Richli, W.
289 R., Savin, M., Waugh, M. K. and et al. (1999) Adenovirus-mediated p53 gene
transfer in advanced non-small-cell lung cancer, J Natl Cancer Inst, 91, 763-71.
Takayama, T., Shiozaki, H., Doki, Y., Oka, H., Inoue, M., Yamamoto, M., Tamura, S.,
Shibamoto, S., Ito, F. and Monden, M. (1998) Aberrant expression and
phosphorylation of beta-catenin in human colorectal cancer, Br J Cancer, 77, 605-
13.
Tan, B. T., Wu, L. and Berk, A. J. (1999) An adenovirus-Epstein-Barr virus hybrid vector
that stably transforms cultured cells with high efficiency, J Virol, 73, 7582-9.
Templeton, N. S., Lasic, D. D., Frederik, P. M., Strey, H. H., Roberts, D. D. and Pavlakis,
G. N. (1997) Improved DNA: liposome complexes for increased systemic delivery
and gene expression, Nat Biotechnol, 15, 647-52.
Tetsu, O. and McCormick, F. (1999) Beta-catenin regulates expression of cyclin D1 in
colon carcinoma cells. Nature, 398, 422-6.
Thomas, C. E., Schiedner, G., Kochanek, S., Castro, M. G. and Lowenstein, P. R. (2000)
Peripheral infection with adenovirus causes unexpected long-term brain
inflammation in animals injected intracranially with first-generation, but not with
high-capacity, adenovirus vectors: toward realistic long- term neurological gene
therapy for chronic diseases [In Process Citation], Proc Natl Acad Sci USA, 97,
7482-7.
To whin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from
polyacrylamide gels to nitrocellulose sheets: procedure and some applications,
Proc Natl Acad Sci USA, 76, 4350-4.
Trono, D. (2000) Lentiviral vectors: turning a deadly foe into a therapeutic agent. Gene
Ther, 7, 20-3.
290 Tseng, W. C., Haselton, F. R. and Giorgio, T. D. (1999) Mitosis enhances transgene
expression of plasmid delivered by cationic liposomes, Biochim Biophys Acta,
1445, 53-64.
Tsujii, M., Kawano, S., Tsuji, S., Sawaoka, H., Hori, M. and DuBois, R. N. (1998)
Cyclooxygenase regulates angiogenesis induced by colon cancer cells [published
erratum appears in Cell 1998 Jul 24;94(2):fbllowing 271], Cell, 93, 705-16.
Tsukada, K., Church, J. M., Jagelman, D. G., Fazio, V. W., McGannon, E., George, C. R.,
Schroeder, T., Lavery, I. and Oakley, J. (1992) Noncytotoxic drug therapy for
intra-abdominal desmoid tumor in patients with familial adenomatous polyposis.
Dis Colon Rectum, 35, 29-33.
Turcot, J., Despere, J. and Pierre, F. (1959) Malignant tumours of the central nervous
system associated with familial adenomatous polyposis of the colon: report of two
cases. Dis Colon Rectum, 33, 465-468.
Urashima, M., Suzuki, H., Yuza, Y., Akiyama, M., Ohno, N. and Eto, Y. (2000) An oral
CD40 ligand gene therapy against lymphoma using attenuated Salmonella
typhimurium. Blood, 95, 1258-63.
Valery, C. A. (2000) In Third International Cancer Gene Therapy MgefmgFaculte de
Medecine Pitie-Salpetriere. van der Houven van Oordt, C. W., Smits, R., Schouten, T. G., Houwing-Duistermaat, J.
J., Williamson, S. L., Luz, A., Meera Khan, P., van der Eb, A. J., Breuer, M. L.
and Fodde, R. (1999) The genetic background modifies the spontaneous and X-
ray-induced tumor spectrum in the Apcl638N mouse model. Genes Chromosomes
Cancer, 24, 191-8. van der Luijt, R. B., Khan, P. M., Vasen, H. F., Tops, C. M., van Leeuwen-Comelisse, I.
S., Wijnen, J. T., van der Klift, H. M., Plug, R. J., Griffioen, G. and Fodde, R.
(1997) Molecular analysis of the APC gene in 105 Dutch kindreds with familial
291 adenomatous polyposis: 67 germline mutations identified by DGGE, PTT, and
southern analysis. Hum Mutat, 9, 7-16. van der Luijt, R. B., Meera Khan, P., Vasen, H. P., Breukel, C., Tops, C. M., Scott, R. J.
and Fodde, R. (1996) Germline mutations in the 3' part of APC exon 15 do not
result in truncated proteins and are associated with attenuated adenomatous
polyposis coli. Hum Genet, 98, 727-34. van der Luijt, R. B., Vasen, H. F., Tops, C. M., Breukel, C., Fodde, R. and Meera Khan,
P. (1995) APC mutation in the alternatively spliced region of exon 9 associated
with late onset familial adenomatous polyposis. Hum Genet, 96, 705-10.
Vasen, H., Bulow, S. and "The Leeds Castle Polyposis Group (1999) Guidelines for the
surveillance and management of familial adenomatous polyposis (FAP): a world
wide survey among 41 registries.. Colorectal Disease, I, 214-221.
Vile, R. G. and Russell, S. J. (1995) Retroviruses as vectors, Br Med Bull, 51, 12-30.
Vogelstein, B., Fearon, E. R., Hamilton, S. R., Kern, S. E., Preisinger, A. C., Leppert, M.,
Nakamura, Y., White, R., Smits, A. M. and Bos, J. L. (1988) Genetic alterations
during colorectal-tumor development, J Med, 319, 525-32.
Wade, N. (1980) UCLA gene therapy racked by friendly fire [news]. Science, 210, 509-
11.
Wade, N. (1981) Gene therapy caught in more entanglements [news]. Science, 212,24-5.
Wagner, E., Zatloukal, K., Gotten, M., Kirlappos, H., Mechtler, K., Curiel, D. T. and
Birnstiel, M. L. (1992) Coupling of adenovirus to transferrin-polylysine/DNA
complexes greatly enhances receptor-mediated gene delivery and expression of
transfected genes, Proc Natl Acad Sci USA, 89, 6099-103.
Wagner, R. W. (1994) Gene inhibition using antisense oligodeoxynucleotides. Nature,
372,333-5.
292 Ward, M., Richardson, C., Pioli, P., Smith, L., Podda, S., Goff, S., Hesdorffer, C. and
Bank, A. (1994) Transfer and expression of the human multiple drug resistance
gene in human CD34+ cells. Blood, 84, 1408-14.
Wasan, H. S., Novelli, M., Bee, J. and Bodmer, W. F. (1997) Dietary fat influences on
polyp phenotype in multiple intestinal neoplasia mice, Proc Natl Acad Sci USA,
94, 3308-13.
Wasan, H. S., Park, H. S., Liu, K. C., Mandir, N. K., Winnett, A., Sasieni, P., Bodmer, W.
F., Goodlad, R. A. and Wright, N. A. (1998) APC in the regulation of intestinal
crypt fission, J Pathol, 185, 246-55.
Westbrook, C. A., Chmura, S. J., Arenas, R. B., Kim, S. Y. and Otto, G. (1994) Human
APC gene expression in rodent colonic epithelium in vivo using liposomal gene
delivery. Hum Mol Genet, 3, 2005-10.
Whitehead, T., Kricka, L., Carter, T. and Thorpe, G. (1979) Analytical luminescence:its
potential in the clinical laboratory.. Clinical Chemistry, 25, 1531-1546.
Wielenga, V. J., Heider, K. H., Offerhaus, G. J., Adolf, G. R., van den Berg, F. M., Ponta,
H., Herrlich, P. and Pals, S. T. (1993) Expression of CD44 variant proteins in
human colorectal cancer is related to tumor progression. Cancer Res, 53, 4754-6.
Wielenga, V. J., Smits, R., Korinek, V., Smit, L., Kielman, M., Fodde, R., Clevers, H. and
Pals, S. T. (1999) Expression of CD44 in Ape and Tcf mutant mice implies
regulation by the WNT pathway. Am J Pathol, 154, 515-23.
Wieschaus, E. and Riggleman, R. (1987) Autonomous requirements for the segment
polarity gene armadillo during Drosophila embryogenesis. Cell, 49, 177-84.
Wilcken, N. and Tattersall, M. H. (1991) Endocrine therapy for desmoid tumors. Cancer,
68, 1384-8.
293 Wildner, O., Blaese, R. M. and Morris, J. C. (1999) Therapy of colon cancer with
oncolytic adenovirus is enhanced by the addition of herpes simplex virus-
thymidine kinase. Cancer Res, 59, 410-3.
Willett, W. C., Stampfer, M. J., Colditz, G. A., Rosner, B. A. and Speizer, F. E. (1990)
Relation of meat, fat, and fiber intake to the risk of colon cancer in a prospective
study among women [see comments], N Engl J Med, 323, 1664-72.
Williamson, S. L., Kartheuser, A., Coaker, J., Kooshkghazi, M. D., Fodde, R., Bum, J.
and Mathers, J. C. (1999) Intestinal tumorigenesis in the Apcl638N mouse treated
with aspirin and resistant starch for up to 5 months. Carcinogenesis, 20, 805-10.
Wolff, J. A., Malone, R. W., Williams, P., Chong, W., Acsadi, G., Jani, A. and Feigner,
P. L. (1990) Direct gene transfer into mouse muscle in vivo. Science, 247, 1465-8.
Woods, D. F. and Bryant, P. J. (1991) The discs-large tumor suppressor gene of
Drosophila encodes a guanylate kinase homolog localized at septate junctions.
Cell, 66, 451-64.
Yang, N. S., Burkholder, J., Roberts, B., Martinell, B. and McCabe, D. (1990) In vivo and
in vitro gene transfer to mammalian somatic cells by particle bombardment, Proc
Natl Acad Sci USA, 87, 9568-72.
Yang-Snyder, J., Miller, J. R., Brown, J. D., Lai, C. J. and Moon, R. T. (1996) A frizzled
homo log functions in a vertebrate Wnt signaling pathway, Curr Biol, 6, 1302-6.
Yost, C., Torres, M., Miller, J. R., Huang, E., Kimelman, D. and Moon, R. T. (1996) The
axis-inducing activity, stability, and subcellular distribution of beta-catenin is
regulated in Xenopus embryos by glycogen synthase kinase 3, Genes Dev, 10,
1443-54.
Zhu, N., Liggitt, D., Liu, Y. and Debs, R. (1993) Systemic gene expression after
intravenous DNA delivery into adult mice. Science, 261, 209-11.
294 Errata
Page 62, line 18 reagent
Page 68, line 10 dependent
Page 93, Figure 7, Lane 11 Empty
Page 137, Figure 12,Lanes 2 to 13 Liver (rather than colonic) 1 ■y 1| iT j>;ïs 1ï > <
All
%
m ê1
mmm
S
M Errata
Page 62, line 18 reagent
Page 68, line 10 dependent
Page 93, Figure 7, Lane 11 Empty
Page 137, Figure 12, Lanes 2 to 13 Liver (rather than colonic)